,%- >=«% .'^■|;

:0 mf:

ÖS." iV

^'^^

\% ■ )% ' j*

r*' 5

^."'r.

.4

.V • 'r iv I'lSf jl-.

:%' "k

.>■ ;s

:i ->■

*\.

M iS- PI :*: fi

m ^ ^5 Ü 5?

m A 1^

JOURNAL

OF THE

COLLEGE OF AGRICULTURE,

IMPERIAL UNIVERSITY OF TOKYO.

VOLUME VIM.

TOKYO.

PUBLISHED BY THE UNIVERSITY 1921-1923.

CONTENTS

No. 1. November. 1921.

I'age NAt:.u, IsAiiuin:— A Otüieliuo-Pliysioloßicftl Study on ihe Foriuatiou of Anthocyiinin aiul

Urown rigmeiils in I'lnnlH. With Place I ami two Ti'Xt-Kigures 1

IkkNci, Si:irrii!(i:— SUulics on ihc (ronetiiw of rUnvor-CoIoura in rorlulaoa graniliflura.

With I'lale li !)3

No- 2, March, 1923.

3L\ltVMo, NoBVKATst) : — Liat of Lupiiiojilora of the Islands Tauegashima ami ^akii»hiiiia.

With Plftle in ia.5

Mastii, Knosni: — The Spermatogenesis of Domctio Mammals. III. The Spermatogenesis

of the lloiise and of the lUbbit. With Vlatc.^ tV-X and om; Text-Figure. . . . •2'17 KoMURo, Hn)Bk>: — Studies in the effect of IJüutgen IJays upon the Development of \'icia

faha. With I'laus XI-XII and one Text-Figure 2.53

No. 3, March, 1923.

Klsiu.novve, KAM.iKicui :- -Contribution.s to the t'oiui)arative Study o£ the So-called Scombroid

Fishes. With Plates XHI-XXXIV and 2« Text-Figures 21)3

INDEX

A. ARllCLES

Pnge

Ikeno, Seniitiko :— Studies on the Genetics of Flower-Colours in rortulncft grandiflora. . . 93

KisHiNot'YE, KAiNtAKic'ni: — Contributions to the Comparntive Study of tlie Socnlled Scom- broid Fishes 293

Kdimr.o. Hideo : — Studies in the Effect of Röntgen Enys upon the Development of Vicin

faba 253

Makviio, NoBtiKATsu: — List of Lepidoptcra of the Islands Tanegashima and Takiishima. . 135

lUsTii, IviTosHi :— Tlie Spermatogenesis of Domestic Jlammals. III. The Spermatogenesis

of the Mouse and of 'the Kabbit 207

KA(iAT, IsABt'Ko:— A Genctico-Physiologlcal Study on the Formation of Anthocyanin ami

Brown Tigmcnts in Plants 1

B. PLATES.

I. A Genetico-Phsiological Study on the Formation of Anthocyanin aTid Brown Tifiments in Plants (I. Naoat).

II. Studies on the Genetics of Flower-Colours in I'ortulaca granditiora (S. Ikexo).

III. List of Iiepidoptera of the Islands Tanegashima and Yakushima (N. MAitUMo).

IV X. The Spermatogenesis of Domestic Mammals. III. The Spermatogenesis of the Mouse

and of the Kabbit (K. Mast'i). XI XII. Sludiis in the Effect of IKntgen Hays uix)n the Development of Vieia faba (H. KoMUEo). XIIl-XXXR'. Cent lil Hit ii lis to the Comprirntive Study of the So-eal1>d Scombroid Fishes ( K. KisniNoi'VE).

C. TEXT-FIGITRES.

Page 1-2. Fcunialion of Anthocyanin and Brown I'igmenls in riants (1. Naoai) 7-45

]. Sjiermatogenesis of the Mouse (K. Mash) 212

J. ICtVecl of Rünlgen Rays upon the Development of Vicia faba (H. Kdiirno). . . . 2(!9

A'/.. Comi'arative Study of the Sc-callcd Scombroid Fishes (K. KisniN(irvK) 311-1G9

Vol. VIII. No. 1.

-^. ^ m ± ^

^ A ^ Jit -

JOURNAL

OF THE

COLLEGE OF AGRICULTURE,

IMPERIAL UNIVERSITY OF TOKYO.

TOKYO.

PUBLISHED BY THE UNIVERSITy.

November 5th, Taisyo X (1921).

PUBLISHING COMMITTEE.

Peof. Z. KIawase, Rwigakuhalcushi, Director of tlie College {ex qßcio). Prof. C. Ishikawa, Ph. D., RigaJcuhakmhi. Pkof. U. Suzuki, Nogakuhahiski. AasiST. Prof. S. Kusaxo, Riguhahakmhi.

CONTENTS.

Page I. Nagai : — A Geuetico-Physiological Stiidy on the Formation of Antho- cyanin and Brown Pigments in Plants. With Plate I and two text- figures 1

8. Ikexo : — Studies on the Genetics of Flower-Coloiu's in PoHulaoa

(jrandiflora. With Plate II 93

All communications relating to tliis Journal should te addi'e.ssed to the Director of the College of Agriculture.

A Genetico-Physiological Study on the Formation

of Anthocyanin and Brown

Pigments in Plants.

By

Isaburo Nagai.

Impeiial AgrictUturnl Exj^riment Station, Jaiinn.

AMtli Plate I aud t^o Text-fisnues.

CONTEXTS

lutroJuction. t . rhysiologiail Study.

1. The Action of Oxiilizing Enzymes on Anthocyanins.

2. The Action of Oxidizing Enzymes on Flavones,

3. Flavone Derivatives as Chromogenic SubstancM of Reddish Brown Plant Pigments

(Phlobaphenes).

4. Eole of Oxygen in the Formation of the Chromogenic Substance and Anthocyanin.

5. A Group of Substances of Unknow-n Chemical Nature As the Chromogcn of Anthocy-

anins and Keddish Brown Pigments. II. Genetical Study.

1. The Mode of Inheritance of Anthocyanin and Brown Pigment in the Awn and Other

Parts of Oryza sativa.

(a) Colour Types of the Awn.

(b) The Cross: Brown x Faint Yellow.

(c) The Cross : Ked x Purple.

(d) The Eelation Between the Colour of the Grain and the Paleas.

2. The Mode of Inheritance of Anthocyanin and Brown Pigment in the Seed Coat of

Glycine sojii.

(a) Colour Types of the Seed Coat.

(b) The Cross: Blue Tinged Yellow x Brown and the Reciprocal.

(c) The Cross: Buff x Black.

[Jotir. CoU. Agric, Vol. Vm, Ko. 1, 1921.]

2 1. NAGAI :

3. An Interpretation of the Kesnlts. i. Discussion. ID. Svimmary and Conclusion. Explanation of the Plate.

Postscript.

Introduction.

The workers in genetics have established the fact that in certain cases the formation of authocyanin pigments is caused by the interaction of a number of definite pigment yielding components which are retained by the separate genetic factore. Neither of these components has the power to produce the pigment unless the complete system is established by their union. We owe much to the Labours of Bateson, Püxnett, Miss Sad^-dees, Miss Wheldale, Bam- and many othei-s on the part of genetics/ and WiLLSTÄrrEE and his collaborators in the field of chemistry who have shown for the first time, the exnct chemical constitution and the inten-elationship of the colijming mattei-s' concerned.

The present paper deals with the result of an investigation carried out in order to discover what rehitiou exists between anthocyauin and brown pigments both of wliich occm- widely in the plant Idngdom and what obser- vation can be made with regaixl to the physiological action of the genes which are analysed by the breeding experiments for the characters in which those pigments are concerned.

I. Physiological Study.

1. The Action of Oxidizing Enzymes on Anthocyanins.

If we accept the view that anthocj-anins are formed by the oxidation of flavone instead of by reduction, and the oxidizing enzymes play an essential part in this change in the hving plant cells, it is uecessai-y to offer the direct evidence to lend support to the xiavr.

When an alcoholic 'or aqueous exti-act of authocyanin which is sHghtly

1. See Wheidale, M., Anthooyanin Pigments ot Plants. 1916.

2. See Pebkis, G. A., and Everest. .\. E , The Natural Organic Colouring Matters. 1918.

A GENETlCO-PHYSIOIrOGICAL STUDY ON THE FORMATION ETC. 3

jicidifiocl to check the fiji'iuation of the ctjluui'lesfs LSf>mer, is mixed with tho »jhition containing active o>dcliziug enzymes, the clmracteristic i-ed colour (limiiiislies gi-mlually and finally becomes pale yellow or practically colourless. The aqueous solutions of liydrogen peroxide and certain other inorganic oxidiz- ing agents have the same effect as the enzyme. Such phenomena have been obseiTcd by Bouffaiid (1902), Kastle (1903), Kastij!; and Haden (1911), Combes (1913), Atkins (191G) and Nagai (1917).'

The reversible change of flavoue to the cfJoured sub.sLance of anthocyaniu- 11 ke natm-e by means of I'educing and oxidizing agents respectively lias been observed by many. Allen (1901)- stated tliat when an acidified (by hydro- chloric acid) alcoholic solution of quercetin was ti-eated wi^i s<xlium amalgam tlio liquid assumed a fine pm"ple colour and on concenti'ation yielded red prisms wliicli dissolved in alcohol and a little alkali forming a gi-een solution, the solution being readily reoxidizod with formation of quercetin on exposui'e to the air. Combes (1913)' obsen-ed a similar reversible change in the yellow pigment isolated from the leaf of Ampelopsis Iiederacea.

In the study of anthocyanin in tho corn flower, Willsatätter and Everest (1913)^ found that when an alcoholic solution of cyanidin was M-armed with dilute hjdrogen peroxide solution, the colour diminished. When it was again warmed on the water bath with the addition of a few drops of (hluto hydi-oclJoiic acid, the liquid Ixjcame j-ellow, and by exti'acting witli ether, Ijeautiful bright yellow crystals were obtained which ^\•llen treated with alkahes, yielded a deep yellow solution. Miss Wheldale (1914)* attempted

1. BouFFARD, A., Action lie I'acid siüfureux sur I'oxydase et sur la matiere coloiante du via ronge. Comp. Bend. Acad. Sei. Paris. 134 : 1380, 1902. Kästle, J. H., A Method for the Dttir- mination of the Affinities of the Acids C'olorimetrically by lleans of Certain Vegetable Coloring Matters. Am. Chem. Jotir. 33 : 4G, 1905. Kastle, J. H. and Haden, R. L., On the Color Changes Occurring in the Blue Flowers of the Wild Chicory, Cichorium intybtis. Am. Chem. Jour. 36 ; 315, 1911. Combes, R., Pas-sage d'un pigment anthocyanique cxtrait des feuilles rouges d'automne nn pigment jaitue continu dans les feuilles vertes de la raeme plante. Comp. Kend. Acail Sei. Paris, 152 : 14,54, 1913. Atkixs. W R. G., Recent Researches in Plant Physiology. 1916. Magai, I , The .\ction of Oxidase on Anthixjyanin. Rot. Mag. Tokyo, 31 : 65, 1917.

2. Allen, A. F., Commer<-inl Organic Analysis. Vol. Ill, Part 1, 440, 1901.

3. Combes, Eoc. cit

4. Wlllstatter, R. and Eveke^t, A. E , Ueber den Farbstoff der Kornblume. Liebig. Ann. 401:189, 1913.

5. Wheldale, M , Our Present Knowledge of the Chemistry of the MendeUan Factors for Flower Colour. Jour. Gvnet. 4 ; 109, 1914.

4 I. NAGAI :

to obtoin the flavone from the authcx-^auiu of Antirr/iinvm by the same mamier jiist quoted, but failed. She obtained oiih- a j-ellowish brown solu- tiou.

HiUatOW aud Gies (1919)' obsen-ed the revei-sible c-olom- changes in the soUition of flavone and anthocyanin isolated from the flower of tuhps by means of nascent hydi'ogen and hydi'ogen peroxide respectively.

The writer obseiTed that the coloured reduction product of quercetin, mj-i'icetiu, apigeuin, and luteolin yielded a yellow solution when ti'eated with hydrogen jjeroxide and in the ctise of the fii'st two, the original colom-s were resumed by fru'ther reduction by means of hydi'ochloric acid and magnesium jwwder. If, ho\\ever, the reduced, coloured solutions were decolourized l>v an excess of hydrogen ^jeroside, the yellow solution so obtained failed to recover the reddish hue by reduction.

The aqueous exti-act of the Aiolet anthoc-yanin n"om the perigone leaf of Iris Kcemjjfcri and the purple one fr'om the leaf of Perillo. nanhmensis were also changed to yellow hj hydrogen peroxide and they yielded again an orange red colom* b}- reduction but no initial liluish hue. \ The red colour of the reduced solution of quercetin, quercitiin, and myricetin were converted to yellow by the addition of an aqueous solution of potassium permanganate, and the fm'ther addition finulh' rendered the solution completely colomless.

The mode of action of hydi'ogen peroxide and the oxidizing enzyme on anthocyanin was studied colorimetiically. One of the difliculties here met with, was the change in hue as the action proceeded. It is naturally to be expected that yeUow colour should increase in depth as the anthocyanin is converted to a flavone like substance and at the point where fifty per cent of anthocyanin is converted to the j-ellow suljstauce, the cc)lüur of the solution becomes about half wa\- between red and yello^^•, namely- orange. When \iolet anthocyanin is used, the change iu hue is so distinct that no dii^ect comparison can be made -tt-ith the standard colom-. With red and orange-red ones, however, the change in hue does not take place in a marked degi'ee up to a certain point, hence the approximate measm-ement Ijecomes possible.

1. Haurow. 15. una Gies, W. J., Experimental Stiulics on TLint Pigments. Columbi.i Univ . Proc. Soc. Exp. Biol. Meil. IG : 8, 1918, Review in Chem. .\bstract. 13 : 2695, 1919.

A GENETICO-rHYSIOLOGICAL STUDY ON THE FORMATION ETC. 5

T<3 illustrate, tlio fi)llowing results will bo ineutioned. Ten gr<ims of tJie fresh petals of scarlet Papavir Rh'jcus^ were exti'actecl with 150 ec of distilletl water. The colonr of the exti-act wius deep red. This was takeu as 100, and 50, 25, auil 12') per cent S(Jutions were prepared. The psroxidase u.sed was prepai'al fi'oiu the pi-essed saj) of hypx'utyls and rootlets of the .soy Ijean see<miigs by precipitating with jilcrjhol. Five cc of the enzyme solution was a<lded to fifty cc of a dilute hych-ogen peroxide solution (0.1 cc corresponded to alx)ut 0.0003 g oxygen). The mixtures were jiut in Lirge test tubes of the (L'ameter 4 ciu and they were kept in the water bath. The tempsratiu-e was kept IjetwGeu 17.0 and 17.5 C. At thlferent intervals of time, five cc of the mixture were pipetted out int<j a test tube and one cc of concenti-ated hydro- chloric acid was added to stop the enzyme activity and to deepen the ciJonr. The standard s<jlution was prepared in the sjxme manner immediately after the enzyme was added. Each portion was then compai-ed witn tha standard solution so prepai-ed by Dubosq's cfJiaimeter and the intensity of colom- of the different poi-tions was expressed as jiercentages of the initial colour ^\hich was taken as 100.- They gave the following result.

T.U3LE 1.

Decolorizntion of anthocynnin by peroxidase. .\. Relative strength of .•mthocyanin = 100 l;. „ = 50.0

C. ,. .. „ „ = 25.0

D. „ = 125

Time in minutes (0	Intensity of colovir {a-x)	fc
0	100.0	0
7.0	100.0	0
40.0	99.06	0.0002t
63.5	93.75	nnnina

1. .\ix«raing to W'illstaitek an.l Wai. i Likbio. Xtxa. 412:231, 1910.) one of the xjigments of a double puriJe-scarlet Tariity is mekocyunin which is a diglucoside of cyanidin.

2. The initial colour corresponded with somewhere between Peach red to Scarlet in Ridoway's Color St.amlard.

I. NAGAI : B.

Time in minutes	Intensity of colour	k
(1	100.0	0
6.5	97.50	0.00101
15.0	90.00	0.00679
39.0	75 01)	0.00738
	C.
II	100.0	0
6.0	93.75	0.01075
15.0	70.88	0.01752
26.5	62.50	0.01773
48.0	56.25	0 01200
C2.0	53.12	00102'1

D.

li	i(;o,o	1)
5.0	81.25	0.04154
14.0	C2.50	0.03323
18.5	55.00	0.03233
36.0	50.00	0.02666
30.5	46.25	0.02535
■lis	43 75	0.01992

Tlie figiu'es iu tliu tliird column nio the vulues of the velocity constsint

Tliev show that the rate

calculated as uuimolecular reaction, /.-== — log. — -

fi — X

of decompositiou in A \\as very slow owing to the relative high concentration of anthocj-anin and the -^aluo of h increased us the retiction advanced. In B, a similar tendency was observed but in D, the value of /.■ rapidly deci-eased as the reaction advanced. Since the strength of the enzyme was the same in .ill cases, the dillbrent ^•tdues obtained indicate that the mode ol action of Ijeroxidase chlTered l)y the relation of the initial concenti'atiou of the enzyme to the suljstivite.

If the values of a—x in A are plotted iigains!; time, ilwy sliow nearly a

A GENETICO-rHYSIOLOCilCAL STUDY ON THE FORMATION' ETC.

; OO

90

80

70

GO

SO

¥0

30

■S 1			T^ r
^						o
>
\	k		^	X,
\	V	X	^	\	\
	\	N.		"^	-^	X
			k^^^^^^^
					^	^^

A

B C

10

20 30

^■0 t

50 60

Text-fig. 1. Gniphs showing the iltcom posit ion of aiathocyanin by peroxiclnse. tee Table 1.

shfiigbt line suggesting that the rate of decompositiou was fipproxiiuately pi'oportional to the time of reaction, heuce the vahie of k calculated for the imimolecular i-eaction iua-eased as time advanced. If, however, the value of log. a — x are plotted against time, the cm-ves of B, C, and D do not form sti'aight lines, while, the values of log. a—x are plotted against log. time, the curves show more nearly the sti-aight lines. This seems to sho^\- that the value of the enzyme was not constant. The active poiiiou of the enzyme de- composed as the reaction proceeded. Two reactions, the decomposition of the enzyme itself and that of authocyanin by the active portion of the enzyme in the S}-stem seem to go on simultaneously, so that the rate of decomposition of the latter cannot be regarded as a simple imi-molecular reaction. The reaction may be uni-molecular only when the specific ratio of the enzyme to the substrate in held and in which the rate of decompositiou of the enzyme may be so slow that its value can be considered as nearly constant, while the decomposition of the substance goes on with the constant rate. Hence we are able to consider the value of a—x as the function of time.

Similar observation was made with laccase. Ten cc of a 0.*) i^er cent

8

I. XAGAI :

laccase Solution' which gave the direct oxidase reaction by phenj-lendiamine, alpha naphtol, and guaiacuni were added to fifty cc of the anthocyanin exti-act which was prejjared from the diied powder of the flower of Lilium tif/rinum by exti'action with dilute alcohol. The mixtiire was kept at the room temperature which varied from 19.0 to 20.0 C.

Table 2.

Decolorization o£ imtbooynidn by laccase.

Tiiuf in minutes	Intensity of colour	/,
(0	(a-x)
0	100.00	0
12	90.63	0.00820
30	83.75	0.00592
50	73.12	0.00626
74	67.00	0.00533
101	62,50	0.00465
•Jul	42.50	iidOii;»

T.\BLE 3.

Same as Table 2, except the concentration of the enzyme which was retluceil to half that of the former.

	Time in minutes	Intensity of colour	k
	(0	{a~x)
.	0	100.00	0
	2	97.31	0.00094
	20	90.77	0.00198
	32	84.27	0.00258
	inn	76.92	0.00202

It was found that certain salts and the sugju-s retarded the a«ti\'ity of the enzyme. One of the experiments gave the following result. Fifty cc of a weak alcohohc (25 per cent) extract of anthocyanin from the violet coloiu-ed perigone of Iris Kcemp/eri to M-liich Avas added glucose (Merck's pure) to

1. The material was IdniUy fumishcil to the writer by Trof. K. Shibata.

A GENEnCO-rHYSHlLOGlCAL STUDY ON THE FORMATION ETC.

9

make up 5 per cent, was mixed witli 0.2 cc of a dilute hydrogen peroxide aud 0.5 cc of tlie enzyme solution which was prepared fiiom the pressed juice of the soy bean seedling twice pi-ecipitated by alcohol. The mixture was kept at 19.0 to 20.0 C. Tlie chauge in the hue fi-om violet to red was obser\'ed as ah-eady mentioned but by aai jKlditiou of sti'ong at-id to the pipetted poi-tion by wliich tlie compai'ison of the colour was made, tho colour wa.s red all alike. Thus :

T.\BLE 4. Influence of glucose on the decolorization of anthocyanin by peroxidase.

Ö per cent glucose	Control (witliout glucose)
Time (0	Intensity of colour	k	Time (0	Intensity of colour (a-x)	7,;
0	100.00	0	0	lüO.ÜO	0
10	97.G1	0.00239	10	90.48	0.00999
20	95.24	0.00244	20	83.81	0.00884
32	90.48	0.00313	40	78.09	0.00589
62	88.57	0.00175	69	76.19	0.00396

The above fact seems to bear some physiological importance. The higher the concentration of sugar or salts dissolved in the cell sap, the weaker would be the action of oxidizing enzyme on anthocyanin. Thus the latter may be protected fi'om decomposition.

As ah'eady mentioned, hydi'ogen peroxide alone decolorizes the anthocyanin and if a small amoimt of finely pulverized animal charcoal is added, the action of hydrogen peroxide is highly accelerated.

The difierent amount of animal charcoal and 1 cc. of hydrogen peroxide wliich was equivalent to 0.014 g. oxygen was added to 5 c.e. of a reduced alcohoUc solution of 0.001 mol. quercitrin. Time requii'ed for the mixtures to be completely decolorized at 14.0 C. — 16.5 C. was as follows :

10

I. NAGAI :

Table 5.

Influence of animal charcoal on the decolorization of nnthocyanin by hydrogen peroxiile.

Charcoal added in gram. (.'I)

No. of mimites required for decolorization (<)

<ll

0.05		29.0	1.45
0.04		34.5	1.38
0.03		42 0	1.26
0.02		G20	1.24
0.015		89.0	134
0.01		140 0	140
Control	ca.	400.0

Wliou the concentration of hydrogen pei-oxide varied and the amount of animal charcoal made constant, at the tem^jerature 20.0 C, the following result ■was olitained.

Table 6.

Same as Table 5.

Charcoal added in gram	Cc o£ hydrogen peroxide added	Concentration equi- Talent to oxygen in g.	No. of minntes required for decolor.
- -O.05 0.05 0.05 0.05 0		1 = 0.&14 1/2 1/4 1;8 1 = 0.014	6.0 11.5 18.5 20 5 ca. 180 0

It seems clear, from the data so far presented, that although the oxidizing enzymes, are univei-sally present in the plant cell, pai-ticuLirly co-existing -w-ith anthocyauin, and although the normal oxygen relation is essential to the formation of anthocyanin in the living tissue as we shall see later, yet these enzymes Lave no direct relation to the formation of anthocyanin from flavone. On the contrary, anthocyanin is converted to a flavone-like yellow suljsüiuce by the action of the oxidizing enzymes.

A GF.NETICÜ-PHYSIOLOOICiU:. STt'DY OX THE FOUMATION ETC. H

2. The Action of Oxidizko Enzymes ox Fiavoxes.'

A\'lien the aqueous or nJwIiolic extracts of the plaut tissues which are lich iu ihivoue, are uiixccl with the freshly prcpared ijressed i^laut-juice oontaiuing the ivotive oxidase, a marked bi-own to reddish browu colour is formed iustautly. The iiiuvt> ilaM)uo the extract contains, the deepar is the colour produced. On standing, a brownish jirecipitate is formed and subsides. Thj production of that colouring matter is produced at the expense of the flavone contained in the exti-act. It can be proved by testing the intensity of the reduction colour of the extract at the beginning and at the end of the experiment. At the end of the experiment, a mai'ked deci-ease in the flavone content can Ije seen by means of its reduction colom-, whenever the brown coloiu'ing matter is formed.

The extract of leaves, twigs, white flowei-s, fruits, and other parts of plants of dLflerent species were examined and iu general, the parallelism iu the depth of the l)rowuisli colour produced b}' tha osidasi and that of the reduction colour of the extract was established. The In-owuish colouring luatter thus formed has its colom- intensified l)y the addition of alkali aud, on tlie addition of acid diminishes or changes to yellow. The colour chauge jast mentioned is vei-y sensitive being performed in an indicator like maimer.

Pure chemical preparations were then tried and it was fouud that certain flavones and flavonols yielded a marked oxidation colour In' the action of oxidizing enzymes. For example, m_^Ticetin, eveu iu a comparatively dilute alcoholic solution, yielded a beautiful red colour immediately after the oxidase was added. The colour, however, was unstable, and changed to brownish red and finally to brown. Quercetin aud luteolin ^-ielded also a deep red colour rapidly clianging to brown. Kasmperol, apigeuin, and tringin on the other hand, showed practically no change. In the former cases, the reduction colour when tested after being acted on by the enzyme, was decidedlj- less deep than that of the control or that of the initial one, while iu the latter cases, practic- iiWy no düTerence ^^■as observed showing that the flavones remained unchanged.

Glucosides gave less characteristic colour than non glucosides. Mjricitriu

1. Tbu till (iccoimt of the investigation will be published liy the author jointly with I'rof. K. Shibata.

12 I. NAGAI:

;ind quereitrin yielded a less bn'gLt red colour tliau cL'd mjiicetin aud qnercetiu respectively.

These oLseiTations indicate that the oxidation colours of flavones and flavonoLs are lai-gely influenced by the chemical constitution of the substance especially the number aud the position of OH group substituted in beta phenyl gi-oup as in the case of the reduction coloiu^.

It is now bej-oud boubt that the action of osid;ises and peroxidases on flavones bears no direct relation to the chemical changes of flavones to authocyanius ; nevertheless the oxidizing enzymes may play an important part in other metabolic changes in the living tissue.

3. Flavone Derivative as Chromogenic Subst.\nce3 of Reddish Brown Pl.vnt Pigments (Phlobaphenes).*

The reddish brown aud brown pigments are -»videly distributed in the plant kingdom, namely in the bark, rliizome, seed coat, dead leaves and iu others, some of them are known as plilobaphenes which were originally the name given by Stähelin and Hofstetteh (1814)- to certain bro\\Taish red substances of unknown chemica,! Constitution isolated from the bark of Pinus sylvestris, Plakinus acerifdia, Bdiila cdba, Ginchomi Ccdisaya etc. According to these authors aud others,' phlobaphenes ai'e considered to be the oxidatiou product of tannins. Hlasiwetz considered that there are two gi'oups of sub- stances wliich may respectively give rise to two gi'oups of phlobaphenes. One of the gi'oups of phlobaphenes to which ' china red ', ' chinava red ', and ' oak red ' etc. belong, yiekls by fusion with alkali, protocatechiiic acid alone, aud the other gi'oup to which ' filix i-ed ', ' ratanhia red ' aud ' chestnut red ' be- long, yields together with protooatechuic acid, phloroglucin. .ilong with them,

1. Nagai, I., On some Keddish Brown Pknt-Pigments. Bot. Mag. Tokyo. 31:1, i;)l7. In Japanese.

2. StXhemx, C. and Hofstetter, J., Chemische Untersuchungen einiger Einden. Liebig. Ann 51 : 63, 1844.

3. Hesse, O , Ucber die humusartingen l?cstandthiUe der Cliina-rinden. luebig. Ann. 109 : 341. 1859. Hlasiwetz, H., Ueber die Beziehungen der Gerbsliuri', Gluooside, I'hlobaphcne iinl Harze. Liebig. Ann. 143 : 209, 1867. Nierbsstexx, M., Beitrag zur Kenntnis der Gerbstoffe II. Berich. d.d. ehem. Gesel. 42 : 353, 1009. See also rEnBiN, A. G. anl Evekest, .V. E., Loc. cit. juigo 436 et seq.

A GE>'ETICO-rHYSIOLOGrCAL STUDY ON THE FOUMATIOX ETC.

i:;

11 ^'oiip i)f snbsttuices such ius nincluriii, ltit«()liii, oatechin, quercetiu niid scopiiriii, also m-lds protocatecliuic add ami plüurogluciu as their deooiu|y«i- tiou producta. So ho sup|x)sed that those pMobapheues which _>-iol(l ])rotoc'ate- chuii- acid aud pMoroghicin are the derivatives of the substance just mentioued. In the epidoruus of leaves aud the psripheiy of the bark, those substances may imdergo oxidation by the contact \\ith air resulting in the formation of phU)bapheno.

^\\vLXEii (1890)" studied thu sklorotie tissue c>f the rhizome of fenis aud came to the conclusion tliat the brown pigment deposited in the membrane is idcnticjJ with plilobapheue and its physiological function was considered l)y him to be the protection of the tissue fi-om the injm-ious eflfect of the humidity of the substi'ate under which these plants gi'o^v.

Some of the brown pigments found in natm-e are hardly soluble iu common organic solvents like ether, alcohol, benzine, aud acetic ether, but are readih' soluble iu water and especially in alkaline solutions, >-ieldiug a deep browuisli red to wine red colom*, which In- acid is iustautlj' changed to yellow. The bro\vn and reddish brown oxidation products of flavoues and flavonols as akeady stated iu the preceding chapter, possess similar properties. Since the flavoue derivatives ai"e widely disfciibuted in plants, it seems quite probable to assume that some of them give rise to phlobapheue Ijy oxidation as Hlasiwetz has alresid_y supposed. Therefore, we may regard certain flavone and flavouol deri^•atives as the chromogeu of both authocvanin aud phloljaphene pigments.

We assume that certain relatioas existing among these pigments may 1« B Dine what as follows :

Chiomogenic substance	Initial cliange	Siibbojuent changes	Product
Certain flavones flavonols ami their glucosides	reduction oxidation	formation of com- plex with salts ~ condensation & polymcrizution	anthocyanins of different hues. phlobaphcnes

1. ■\Valtek, G., Ueber die braunwandigen sklerotischen Gewebeelemente der Fame, mit Ijesonderer Ucriicksichtigimg der sog. " Stützbündel " Eussow's. Bibüotheca Botanica. Heft 18:21, 1890.

2. See SmBATA, K., Shibata, Y. and Kashiwagi, I., Studies on .Inthocyanins. Color Vari.a- tion in Anthocyanins. Joiir, Amer. Chem. Soc. 41 : 208, lül!).

14 I. SAGAI:

4. II iLE (.'F OXYGEX IX THE DEVELOPMENT OF THE ChEOMOGENIC

Substance and Anthocyanin.

It is akeacly kuown that the formatio:i of anthoc3-amii is suppressed when the uijimal oxygen relation is artificially fhecked in the Hving plant.'

The writer observed that the hypocotyls of the ssedh'ng of the buckwheat remained white as long as they were kept in the dark, but when exposed to the day light, a deep red coloiu' developed. If they were kept in the glass chamber in which the air was replaced by hydrogen gas, they did not form tlie pigment even when expossd to the day hght. The chromogenic sub- stance could be detected from the colouiiess samples. The alcohohc exti-act gave a distinct red colour by reduction by means of hydrochloric acid and magnesium powder."

The yoimg seedhngs of certain varieties of soy bean form a deep pm-ple anthocyanin pigment together with chlorophjll iu the elongated hypocotyls a few days after germination, if they are exposed to sti'ong sun light. If, how- ever, tha normal supply of air is checked, the pigment does not develop as in the case with the buckwheat. The following expeiimeut shows clearly the above relation. Tlie j'oung seedlings of a vai-iety of yellow cotyledon which were devoid of both pigments, were placed under the following c-<Hiditions :

Lot A. A small, tightly fitted glass chamber in which the air was replaced by hydi'ogen gas which was generated by Ivipp's appiu-atus and washed by the solution of potassium permanganate once. The chamber was dipped in water.

Lot B. Same as A, lint the chamber was kept iu the air.

Lot C. The chamber was simply closed up.

Lot D. The chamber opened, as conti'ol.

AU the chambers were kept in a glass case which was placed near the

1. Emery, M., Sur les viiriations ile 1' emi tlnni les iierinntliiB. Bull. Soc. Cot. Fnmee. 36 : 322, 1889. Katic, I). L., Beitrag zur Kenntnis iler Bililimg des roten Farbstoffs (.\atliocyim) in vegetativen Organen der Phanerogamen. Iniiug. I)is.sert., Halle. 1905. Citexl in Wheldale, M., .\nthocynnin Tigments of Plants. 1916.

2. See also Miege. E., Kecherches siir les prinoipales especes de Fagopynun. These de Doctoral de 1, Univ. Paris. 1910. Cited in Combe?, B., Kecherches biochemiiiues expcrimentales sur le röle i>hysiologiquc iles glucosides ehcz les Vegetaix. Eev. General, d. Bot. 30 : 89, 1918.

A GENETICCi-PHYSIOLOGICAL STUDY ON THE FORMATION ETC. 15

wiiuliiw to receive direct suulight. Tlu-oe clays after the experiment was set up, clilor>)pliyll clevelops<l in the sec{lllnfi;s in C and I> ; the deep purple pigment w;i3 also formed in the latter. None of the pigment were formed in

A. On fourth day, a slight purple colour was observed in C but in A and

B, it fjiiled t<) appsar. lu tlie same day, the chamber B was opened to allow normal air. Two days later (on the sixth day) the saedlings in A remained stiU without the pigment. In B, chlorophyll was found in the cotyledons but no pnrple pigment in the shoots. In C, the development of the pui-ple pigment was still feeble, whereas in D (control), all the seedlings were deeply colom-ed.

It is a well known fact that the leaf scale of the onion bulb becomes yellow on exposm-e to light. It is chiefly due to the formation of quercstin.' If, however, the bulbs of which ali'eady colom-ed scales were removed, were kept in the closed chamber filled ^^•ith hych-ogen gas, the formation of flavone was inhibited. The bulbs were cut in halves and the coloured scales were removed. The halves were kept in the closed glass chamber in wliich the air was replaced by hydi'ogen gas and the other lialves were kept for the control experiment. They were kept for sixteen daj-s dming which the gas was renewed once. Neither yellow nor gi'een pigment was found except in the control specimens. When the chambei-s were opened, the bulbs were tiu-gid, but liecame soft immediately after the air was let in. An equal weight of the scales was taken fi'om the ti-eated and conti-ol samples and exti-action was made with equal volmes of a weak alcohol. Tlie exti-a«ts so prepared, were reduced by means of hych'ochloric acid and magnesium powder in the usual manner. They gave the following flavone reactions :

Ti-eatetl Ti-ace of pink coloiu-.

Conti-ol lied.

It showed that normal aü" is essential to the development of flavone in the leaf scale of the bulb of onion even when light is amply supplied.

Tlie bulbs of AUium Ledchourianum, a common weed in certain pai-ts of Japan, ai-e white when they ai-e gi'own in the gi'ound, but they ai"e dug out

1. Perktm, a. G. ftncl HrsniEL, J. J., Occiurence of Quercetin in the Outer Skin of the Bulb of the Onion (.tUiiim cep.-i). Chtm. Soc. Trans. C9 : 1295, 1896.

16 I. NAG.U ;

aud exposed to day light, a deep violet-red coloiu' develops. The colom-ed scales were removed and the white portions were kept in the same manner as in the previous experiment ^vith the common onion bidhs. No red pigment was formed during the experiment Mhich lasted for two weeks. The bullös in the conti'ol Ijecame deep gi-een and the treated ones were white. Alcoholic exti-acts were prepared, and the flavone content was examined. They showed :

Ti-eated No reaction

Control Trace of pink colom-

Here the flavone was formed in exti-emely shght amoimt even when the light and the air relation was normal.

5. A Geoit of Si.tbstanc'es of Unknown Chemical Nature

A.S THE ChEOMOGEN OF AnTHOCYANINH AND

Reddish Brown Pigments.

A body of evidences accumulated by the diflerent iuvestigatoi-s show' that a gi'oup of chixjmogenic substance is present in plants which gives rise to anthocyanin-like pigments as well as the brown pigments.

Wolff and Kouchermann (1915)- observed the presence of a chromogenic substfince in a number of plants which are sensitive to the action of Laccase .yielding the brown pigment. The phenomena observed in iodine coloiu- tests are always pi-eceded by the action of a laccase. The authore considered that the chi-omogens are of the same kind in diflerent plants and the brown pigments Avhich are formed in various plants or organs might lie regarded as products of oxidation. Shibata' obtained the chromogenic substance fi'om a

1. Mai.vezin, Ph., Sur 1 'origine de la coIoiu ties misins rouges. Comp. Eencl. Acad. Sei. Taris. 147 : 318, 1908 ; Laboede, J., Siir 1 'origine de la matiere colorante des raisins rouges et mitres orgones vegetans. Ibid. 146 : 1411, 1908 ; Dezani, S., Le sostanze cromogene dell' uva bianca. Staz. sper. agr. ital., Modena. 43 : 328, 1910 (cited in Atkins, "\V. E. G., Kesearclies in Plant Physiology. 1916) (also in Jour. Chem. Soc , Abstract 100 : 223, 1911.) ; Keegax, P. Q , Tlic Chetaistry of the Flower Pigments. Chem. News. 107 : 181. 1913 ; Tswett. M., Zur Kenntnis des vegetabilischen Chama-Ieons. l?er. d. d. bot. Gesell, 32 : Gl. 1914; Tswett. M.. BeitCige zur Kenntnis der Anthocyane. Uebcr künstrichcs Anthocyan. Bioch. Zeit. 58 : 225. 1913.

2. ■\VoLFF, .T. und EoDCHEKiUNN, K., Sur les proxirietes d'un chromogene universellement repandu dans les vegetaiux. Comp. Eend. Acnd. Sei. Paris. 161 : 399, 1915 : 'WoLrF, J., PhenomC-nes d" oxidation et de reduction portant sur les chromogenes des vegetaiuc. Had. 160 : 716. 1915.

3. Shibata, K, Bot. Mag. Tokyo. 31 : 1919. A brief note in Japanese.

A (.KNETICO-PHYSIOLOGICAL STUDY ON THE FORMATION ETC. 17

iiuiuIx-T of plants wliicli Woklwl a deep reJ colmir ou lioatiug ^^it]l Inclrn- cliloric iifkl mill with oxidfise n bi-owu t<i nxklisli foloui'. The red coloiuiug; luatter obtrtiued by heating \\ith the acid is chauged to l)hic by au alkali like authocyauiu. The chromogenic substauce is solulile in alcohol aud iu ether, and with ammonia jiclds a deep yellow eok>ur as ol)ser\ed with flavone I Hit 1 )y reductiou jields uo colour. It is very seusitive to the action of oxidases forming a brown substance of wliicli the colour is intensified by alkali aud chauged to yellow or yellowish brown by acid. The plant extract which con- tains both this sulistance aud certaiu kinds of flavoues shows a characteristic i-ed colour by reduction as well as wheu heated with hydrochloric acid. Shi- B.\TA considered that the substance might be regarded as a colourless autho- cyauiu. MoiiEAix (1914)' considered it pTOp9r to rank aloug with red, viiilet aud blue pigmeuts designed as authocj-auius, the colourless compouuds which are inseparable from them aud which are ah\ays foimd in the celLs as earlier or later products, l)eing closel}- related to them as regards chemical comj^wsi- tif)U find as haviug iu common with them a mitochoudrial origin.

We do not know as yet the chemical uature of the si:bstauce iu questiou l)ut the similarity in ceiiain properties exhiljited by the extract fi'om a uumlier of plants, suggests tliat a closely allied substauce maj- be present wideh' iu the plant kingdom aud it may give rise to certaiu auth'jcyauins aud the reddish bn)wn pigments. The Ijeariug of the fact on genetics is hardly to be overlooked, for we are uow able to locate aud to approximate the chromogeu iu the part of a plant by the test for flavoue aud the substauce under discus- sion. Iu the following pages, the former will be uamed, for the sake of cou- venience, chiümogeuic substauce F, aud the latter clu'omogeuic substauce P.

A number of species of plants especially the cultivated plants, were ex- amiued for these clu-omogenic substauce. The method employed ■\\as as follows. To each gi'ara of the fresh material, ten cc of a weak alcohol were added and exti'acted on the water bath. Usually three to five gi'ams of the materials were taken. Five cc of the extracts were reduced with oue cc of concentrated hydi'ocliloric acid aud magnesium powder for the chromogenic sulistance F, and another five cc were simph' boiled with the acid which w.as

1. JIoBEAüx, F., Ij' origine et les transf irmations des l)^o<^^üts anthoej-aTÜnues. ISiüI. d, I oc. 15üt. France. Gl : 390, 1914.

18 I. XAGAI :

added iu the same proportion as before, for the chromogenic substauca P. After the treated extracts were cool, the coloiu^s of the extracts were compai-ed with those of the standard coloiu-s and recorded. The staudai'd colours were prepai'ed iu the following manner.

Twenty five cc • of quercetiu dissolved iu absolute alcohol, wer j reduced wlthfive cc of concentrated hydi'ochloric acid and about 0.5 grams of magnesium powder.

Colonr scale	Concentration of quercetin
I	I: 1,000
II	I: 2,Ono '
ni	I: 3,000
IV	I: 5,000
V	1 : 10,000
VI	1 : 20,000

Thus the relative value of the cluomogen content iu the material was approximately determined. The result of a survey established the following fact.

The cln-omogeuic substance P cau be detected iu dillerent parts of plants, i.e., the leaf, stem, shoot, rhizome, bark, wood, wliite petals, perigoue, seed coat, mesocai"p, stigma etc.

It occixrs quite iudepandently or in company ■n-ith the chromogenic sub- stance F. Even iu the same plant, the distribution of the two chromogens is quite chstinct in diflerent organs.

Light seems to have no tlirect relation to the distribution of the chi-omo- genic subs':a.nce P unlike the wise of the chromogenic substance i^ (flavones).' For, the uudergroimd parts and the interior portions of the upper ground tissues of many plants contain a considerable amount of the chromogenic substance P. The bark of young twigs, the seed of imiujitnre seeds which ultimately become brown, red or black when fiilly matiu-e, the young fraits and berries of mauy of the cultivated fnüt trees are especially rich in the clu-omogenic substance P. Thus, for example:

1. Shtbata, K. Kagai, I. anil Kishida, JI , The Oocnrrenoc anil Physiological Significance of Flavone Derivatives in Plants. Jour. Biol. Chem. 28 : 93, 191G.

A GENETICOPHySIOLOOICAL STUDY ON THE FORMATION ETC.

19

Table 7.

Chroiuogen content of tlic |ilant extracts.

Nlllllf nt pliiJll	I'art exnmincil	Kelative value 1'	of chromogens y
Cryptomeria japonkn	Green leaf	I	IV
rinus parvifolia	Green bark Wood & i>itli (young twis) Green leaf	I+ + + 1 IV 1 +	IV + + V
»	-liark	I	V
I'lataiuts occideiUfilis	Hark	1+ + +	w-
»» >»	■Wood	IV	-
At\iculri.i Oirbiivitii	Bark	1+	+
	W. .M,l	TV	-

In the immature gi'een seed of difi'erent legumes, the chromogen can readily he detected, so that it is possible to predict whether the seeds may be coloured or n(jt in adnlt condition. S;) far the writer has found, the colourless or wliite seeds show practically no chromogen reaction when they arc stiU young, tlie relation of the pigment to the chromogen being quite well marked.

Table 8.

Chromogen rontent of the extracts of green seed of the legnminoiLs plants.

Xamc L'f plnnt	I'.irt txamined	Colour of seed when ripe	Kelative value of chromogens r p
Indijofera pseuäotbuiorUi	Peed	Brown	I	_
VicUi sathn	,.	.,	n+	-
Pisum salimm	,.	„	1+	+
Phaseolus vnlijai-'is	„
'• Kotcnnshi "	„	White	—	—
" Ohtenashi "	„	„	—	—
" Kianeko "	'•	Yellow pie<l	IV	+

1. + sign denotes the deei)er colour tlinu that of the colour scale stated. The sign alone denotes the trace of colour or below VI.

20

I. NAGAI :

Name ot plant	Tart examined	Colour o£ seed when ripe	Eelative value o£ chromogens /• F
" Golden Wax "	See.1	Black	- I. ! +.
" Chosen "		Eeddisb brown	II +
" LongfeUow "		Brown mottled on light yellow	II +
" Kumamoto-Ingen "		White	^ ~
" Bkck A'nlentine "		Black	II
" Canadian Wonder "		Reddish browia	II
" Birma-Ingeu "		Black mottled on brown	i^
Phaseolus raüudus var. mireits
Unnamed		White (i ale biiff)	-	-
" Shiro-Adzuki "		,.	-	-
Unnamed		buff	I- (;')'
,.		lied	It IV
" Maru-Ba "		„	1 +	IV
" AVase-Dainagon "		„	1 +	IV
" Jladara "		Black fleck., on red		IV
" Kensald "		Eed	I IV
" Yogore "		Dark fleck;> on b\iff	11 -
" Midori "		Greenish grey	I i (")
" AVase-Otsuhu ■■	Dark red	I T\

As we Luve jiist seeu, the clu'omogeuic substance P is pleutifiiUy foiiud iu the exti-acts of the immature coloured seed of Phuseolus vulgaris, but very scai'ce iu or uearh" devoid of the chi-omogeiiic substance F. In the leaf, the reverse is the case.

T.VBLE 9.

Chromogen content ot the leaf of riiriseohis iii'.ijarh.

Nuaie ol variety	Colour of flower	Colour ot Icat	Chromogen r F
" Kotenashi " " Ohtenashi " " Kianeko "	White Cream	Yellowish green	-	11 i\'

1. Oranqe colour.

A GENETIGO-rilY.SIOLOCaCAL .STUDY ON THE FultJ[ATI(JN KVC.

21

"Snmc of vnriely	Colour of flower	Colour of leaf	Chromogtn 1' P
"GoMm Wax"	Tale pink	Greenish yillow	+	IV
■' t'liostn '	.,	Green	—	m
" Longfellow "	White	»»	—	III
" Knmamoto-Ingca "	Cream	»»	-	IV
" \ilnck Valtn'.iue "	Pule pink	»»	+	IV
" Cnnndinn Wonder "	rink	Greenish yellow	—	in
•• Kinimn-Ingt-n "	••	"	-	III

Unlike the c:ise of the above, V)otli clu'oiiiogonic substances co-exist in tlie leaf of certain fiiiit trees.

Tablk 10. Chromogen content of tlie leaf of certain fmit trees.

Name uf [ilaut	Cliromogen I'	Chromof^en F
Graiics (^Vitis)
•• .\<liron lack "	II	I
" Bacon "	1 +	1
•• Brighton "	I	I-
" Catawba "	I	II
•• Champion "	I-	II
" Concord "	I	II
"Hartfor.l Prolific "	II	n
•• riervert "	It-	If
•' Koshn "	1 +	I
" Ives "	1 +	11 +
'• Lady Washington "	I	II +-
■• Sweet Water '	I+ +	I
Sand Pears (i'irii.s seroihvi)
•• Nijiiseiki "	I	I
" Kozo "	II	II
" Chojiiro "	11	II
.\pples (Mulvs xybestr'ix)
" Iwai "	ni	HI
"Jonathan" ("Kogyokn")	III	1+

22

I. NAGAI

Name of plant	Chromogen P	Cliromogen F
"Eawles Janet" ("Kokko") "Smith Cider" (" Kyiigyokii ") "King of Tempkins Country (" Hinokoromo")	m III ni	I n n

If ^\■as uoted tliat the reduction colour of the extract of the leaf of the pear aud that of the apple was somewhat different ; the former was orange red, the latter more scarlet red, suggesting that different flavoues might be present in them.

In the leaf of 3Iorvs al/>a, which is \evy impoiiaut in sericulture, the chromogenic substance F was found but uo- the other, while in the le;if of Irifi Kampfe)-!, the reverse was the case, and in the latter, the clu-omc^en content in the leaf was not con-elated with the coloiu- of the p.n'igone, some of which are deeply coloured.

Tablk 11.

The chromogen content of the leaf of Mortis allxi und Iris Ktinnpferi.

Käme of jJunt	Chromogen 1'	Chiomogen F	Bemivrks
SIoi'us alhfi
" Jlishima "	-
" Akagi "	-
" Ko-so "	-	III
" Eo-so " Seedling	-	IV
" Furisode "	-	IV
Iris Ka:mjfen
No. 1	I	_	The colour of perigone solid hyacinth purple.'
No. 2	II	—	Flecked violet purple
No. 3	III	—	Solid Eood's violet
No. 4	in	-	Flecked violet purjile
No. 5	I	(V)	Self p.'uusy violet
No. G	I	<J)	White

1. Nomenclature acconUng to liDiavAV, K. e'olor Mim.Uirds and Color Numt-ncL-iUin-, liilii.

A GENETICO-rHYSIOLOOICAL STUDY ON" THE l(iH>rAnnN ETC.

23

The skiu of the youug, gi'ceu gi'iipas was füund to Ije very rich iu the chromogeuic siiljstjiuce P l>iit it coutuius ouh' u ti-aco of the chruiiKjgeuic siibstiince F. It is of interest to find that the "white" varieties, so far examinetl, coutaiuetl the clu'omogenic substance as luiich as that showii by the coloured varieties. Eveu in the deep Ijlack variet}-, the reduoticui colour of the extract of the green skiu was only faint, so it seems highly probal)le that the authocymiin pigment in the skiu of the gi-ape may lie formed chiefly from the chixjmogenic substance P rather than the clu'omogenic substance F. According to WillstXtter aud ZollesTiER (191.5)' the anthocyanin of gi-ape skiu (North Itnliau or hothouse) were seuidin .tud PPuiu which are the methyl ethere of dclphiuidin and delphiuiu l•os^5ectivol^■. The chemical investigation of the chi-omogenic substances iu the gi-ape skiu is inviting, for it may cleai" the relation of anthocyanin to the clu'omogenic substances P aud F.

Dezani found two kinds of the clu'omogeuic substances in the white grapas of which one only is precipitixted Ijy lead acetate. By the action of hydro- chloric acid, coloming matters are obtained which are analogous to the ajuocy- anius. The conversion of these substances into colouring matter is due not to oxidatiou, but probably to hydrolytic scission ■\\ith simultanious formation of a reducing substance. In the residue fi'om the clu-omogeuic sulistauces there are other substances which give a red colour with alkali. The result obtained by BteDRiCK and Anthony'' shows that (1) " wliite " is a pure, colour, namely "white" X "white" gives ouly "white": and (2) "white" is recessive to both Ijlack aud red.

Table 12.

Chromogen content of the sldn of green gmiies.

Name of variety	Colour of skin when fully ripe	Chromogen P.	Chromogen F.
" Ives Seedling " " Bacon "	Black	I I	VI VI

1. AVrLLSTÄTTEii, E. anl Zollinger, E. H., Ueber die Farbstoffe der Weintraube u. der Heiderbeere. Liebig. 408:83, 1915, and 412:195, 1916. See also Dezani, S.. loc. cit.

2. Hedeick, U. p. and Anthont, K. D., Inheritance of Certain Characters of Grapes. N. Y. Agric. Exp. Station. Technical Bidl. 45. 1915, pp. 19.

24

I. XAGAI :

Name of variety	Colour of skin when fnlly rii-ie	Chromogen P.	Chromogen F.
" Concord "	Black	1 +	VI
" Black Hamburg "	,.	1 +	M
" Hervert "	»*	I-l-	VI
" Bryan "	,.	1 +	VI
" Delaware "	.,	If	VI
" Cthelo "	Ked	1 +	VI
'■ Bryant "		I	VI
•• Lady Washington "	White	Ih	VI
" Vergonnes "	„	II f	V
'• Moore's Diamond "	„	1 +	VI
"Niagara "		1 +	VI
"Goldfn Champion"	Amlier	I	VI

"Bell," "Esther," "Eatou" (whites), "Highlaud," " Hartford Prolific " (blacks), and " Iowa " (red) showed likewise the marked coloiu- reactiou of the ohroinogeiiic substance P.

Apples, peal's, oranges, Kaki fi'iiits, strawberries, bananas and other fruits and vegetables Avere examined, the results of which were listed and given in Table 13.

In certain plants, the chromogen reaction failed when tested just before the formation of anthocyanin. The leaf scale of the bulb of Liliuta tujrimim is devoid of anthocyanin when it lies underground. But it becomes pmiJe in exposure to sunlight. Even a few hoiu:« exposm-e causes the purple spots to appear on the surface of the yellowish white leaf scale, and the coloured area extends graduallj* to the entire scales, within a few days. The alcohohc extract of the suitable material, however, shows practically no reaction of either chromogen.

The potato tuber is another example of this kind. Two kinds of white tubei-s are known. One is such that the tubers ai'e devoid of anthocyanin as long as they are in the ground but by exposm'e to sunlight, they become deep purple. The other is such that, even when expossd tt) the light for a long time, no anthocyanin is produced. The exti-act of both kinds of white tubers showed a very feeble reactiou of ths chromogens. It seeins that the anthocyanin maj' be formed so rapidly from tha raw material in these instances

A GENETICO-rHYSIOLOGIC.\L STl'DY UN THE FOIOrATION ETC. 25

tliat there may Ije no appreoiiiblo lunoiint of tlie cLromogcuic substance ae- cnijiidated to sliow a ilefiiiite colour reactiou.

Tlie body of evidence so far rejxirted seeius to point to tlie following conclusions.

In a number of plants, autbocyaniu and tlie br(jwu pigment (pbloba- phene) vnn 1)6 traced t.) their rospactive chi-omogenic substiince pre^•ious to the formation of the pigments. Both pigments can be formed fi-om the same cliromogenic substance by the action of a number of complementary pigment- j-ielding agencies. The clu-omogeuic substances can te identified as belonging to two giHJups of substance with respect to certain colom* reactions, one of which is designated as chromogenic substance F, (certain flavones and flavonols), and the other as clii-omogenic substance P of which the chemical natiu« is luikuown.

The formation of brown pigment includes at least the following cases.

1. It is cliiefly due to the oxidation and subsequent changes of the chromsgenic substance F. Example : the awn of Onjza scdiva as we shall see Liter.

2. It is diiefly due to the oxidation and subsequent changes of the chixjmogeuic substance P. Example : the seed coat of the legumes {Pkiseolus vulgaris, Pisum sativum. Glycine sqja etc}.

3. It is cheifly due to tha oxidation and subsequent changes of the chromogenic substance F and P. Example : the barks of many trees ( Cnj- ptomcria japonica, Pimts parvi/dia, Plcdanus occidentalis etc).

Besides those chromogens, tannoids, and Carotinoids may play a role in the pi-oduction of the reddish and yellowish lirowu pigments.

T.iliLE 1.3.

Showing the chroinogen content of the i^lant extracts. Designations : Peri. = iieriphernl tissue. Int. = Interna! tissue.

( + ') sign w-ithout the number of the class in the colour scnie, denotes the presence of the colour but below the lowest class in the scale. ( + ) sign with the nnmemi denotes the colotir somewhat deeper than the senle indicated by the numeral. ( — ) sign designates hkewise the colour below the scale.

26

I. NAGAI:

( X ) sign designates the presence of n distim-t colour which

differs from that of the coloiir scale.

K=reJ, 0 = orange, OR = orange red, B = blue,

V = violet, M = iu.igenta. Y = yellow, G = green,

YG = yellowish green, Br=brow-n, BR = brownish red.

Mame of plant

Part examined

Chromogtn F F

Kemarks,

Gymnosperms.

Cryptomma japorica

J» f,

rinus deiisiflora

Lar'tx UptolejAs

Ficea ajanensi^

Thujopsis do'.ahrata

J» »■

Angiosperms

Alismataceaj

Sagitiaria sar/UHfolia var. longiloba f. sinensis

Arncete

Colocasia m.ti'juonun

Lilioceaj Allium Cupn A. fislulosum A. LedebouriarMm FJrythronium denscanis.

Zitium (igrinum

Atnaryllidnceie Ifarcissiis sp.

Lf.
■Wood fc	pith.
Lf.
Bark
AVood &	pith.
Lf.
Wood &	pith.
Bark
Lf.
Bark
AVood &	pith.
Lf.
AVood &	pith.

Bidb. Peri. „ Int.

Ehizome Peri.

Int.

Leaf scnle

Lf.

Lf. scnle

Shoot

Lf.

Lf. scale Peri.

,. Int.

Lf. s<'ale

in- i\ r\' I

VI-

lU

V

I

HI

I

V

m-

V

YI

+

VI VI

VI

+ + +

V

+

X VI

YI

II

Young twig.

Ill I Y'onng twig.

V +

lY

Y

The outer skin is blue.

AYhite tissue.

The outer skin brown.

■\Vhite tissue.

Etiolated.

Etiolatetl part. Green.

A GENETICO-PHYSIOLOGICAL STUDY oX THE FORMATION KIV:.

27

Nnme o£ jiliint	Part examineil	Cliromogen r F	lU-marUs
Karclssus sp.	Lf.	—	\1-
„	Perigone	-	11+	Yellow colour
ft	Corolla	-	n+	Pale yellow
Pioscorerice.-B
Dioncorea Batata	Hhizome Peri.	VI	+	The hkin Ijrown.
»t »1	Int.	Y	—	White tissue.
Iridncene
Belamcanda punctata	Fnut	VI		Green fniit, black when ripe.
Bromelincea;
Anamis satiats	Fniit I'eri.	11 +	II
>» »»	„ Int.	x(Y)	II
Commelinncefe
Commelina commimis	Seed	VI	+	I'r.ripe seed, brown when ripe.
Graminere
Ar/rostii vulgaris	Lf. cV: eiilm.	—	VI+'
/Ueno sativa
" Tresspass"	»• »f	—	IV
" Kolmoen "	t>	—	IV
" Eiice Horse "	M it	—	IV
" Hnilnka "	■» »»	—	r\'
Dactijlls glomeraia	»» •»	+	IV	/
llordenm iiitgare
Spring barley A	.,	-	IV
.. B	»» •'	—	r\"
C	»» J»	-	IV
■\Vinter barley A	,.		V
B	I» »t	+	JX
C			n-
"Golden Melon"	T' »»	—	IV
Triticum
" Gypsy "	I» »•	-	IV +
"Keel Wave"	»* 1»	—	rv+	'
" Silver Sheef Longlessy "	,.	-	n-+

1. The reduction colour o£ the extract of most of the grasses examined i.s tinged with orange red hence the coloiu can be matched better with the scales which are made Ijy the reduced solution of apigenin or luteolin.

28

I. NAG.Vt :

Xamc of plant

Part examined

Chromogen P F

Eemarkes

" St. Lous Grand Prize "

" Fulcaster "

" Castle's Prolific "

" Imperial Amber "

" Eclips "

" Ptnn. Blue Stem "

" Valley •■

" Harvest King "

" Pool "

" Kiiports Grant "

" Mortgage Lifter "

" Rnml New Yorker "

" Fiütz "

"Jones Mammoth Amber'

" Giant Sfjuare Head "

" lüenilyke "

"Dawson's Golden ChafI "

" Kanreil "

" Turkey "

" Kiilirküf "

" Eikun No. 1 " TrUimin fipcrtwn Zea Maiji

" Koshu "

Unnamed

Sccale c(rea'.e

Setaria italicd

" Honaga "

" Tsiigani-wase "

" Akaho "

" KarasiUiashi "

" Honaga-Siisa-iiwa " " Eda-awa "

" Aka-gara " ■' Bukkiri " ranltum firwinaUaccwn

LS. & iilm.

Seed

Lf. A- cnlm.

Panicle Iji. & I'.nlm.

Panicle

Lf. & evilm.

Panicle

x(Br)

x(B)

(Br)

■i-

+ x(BG)

+

+ x(B)

+ x(B)

IV + III IV + IV + IV + IV + IV

in III

IV

m

rv+

III

ry+

IV + IV + IV IV + IV + IV IV-

in

in

Unripe seed. "White dent.

Young head

V Y'oung head

A OKNETICO-PHYSIOLOOICAL STl'DY OX THE FOItMATION KTC.

29

Name of piniit	Part uxaniiniil	Chromogeu P F	Kemark<.-i>
" Choim "	I.f. * culm.	+	in
" 'Xigiri "	..	+	III
•• liiingoro"		+	II
" Sbinilai-naoahi "	•■	+	III
•• Shiro-hie "	f *•	+	m
•' Scnkokii "	..	+	II
" Oimga-liie "	Heal	—	iv
" Shirobfinn "		—	+
*' Kistn *'	..	—	-
" Sluro-sjxnyokii "		_	VI
" Kebie "			-	Yo".ing head.
" Oso-bic "		ni	Ill	Anthocyanin pre- sent.
" Chosen "	,. ,,	VI	\i	Anthocyanin pre-
Chryia sal in	(See Table 14)			sent.
'• Aikoku "	I.f.	_	V 1
" Biingo "	,,	_	V
" Oba •'		_	VI
" Asaterashi "		_	NT	The awn red.
" Kiiraftisngi "		—	VI	The awn brown
•• Daikkoto "	,,	_	VI	,, ,, ,^
"Uhei"	„	_	VI	,.
•' Akage "	,.		V	,,
•• Cnowftse "	„	_	VI
" Vaniato-chilvjirii "	••	—	VI	Tue awn faint yellow.
" Sekiyama "	„	_	vr	The awn brown.
•• Kamcno-o "		_	\-	Awnless
•• ( Jcnrolni-mochi "		—	\	The awn i)iirple.
■• K:nvabe-m(X;hi "	„	_	V
" Dailiok« "		—	vr	Pwarf jilant.
•• Shild-sbima "	„		IV
Miisnc«c
Jftisa sp. (Banana)	Fnut Peri.	VI	VI	Skin, fiUly ripoil.
"	.. Int.	IV		Fresh, yell'jwi-b

1. The colour scale was prejiared by the flavone isolateil from the leaf of (fysu sniha instead of nuercetin.

30

I. XAGAI :

Name of plant

Part examineil

Chromogen P P

Zingiberacene

Zinrjiher officinalis

Cupiiliferie

Castdnea mitioi

Jiiglandacefe

Jwjlmw Sieboldhnvx

Moracefe

Mortis aVia CannabinaoefE

Cannabis .intica Polygonaeea;

Fagnpijrum vnljare

Jieynoutria japoiiim Bheum Bkapoixticum Chenoixxliaceje

Sphmcia oleraaea Nynyihreaoe.-c Ndwnho niidfera

Cruciferfe Bi-assica campestris var. rapifera B. ola-aeea var. liriills B. o'.eracea var. capitata

B. o'.eracea var. lolri/tls " *» >» »)

Jikaphaniis sativus

" Horyo "

MalvaceiG JJp/iscus syriacns

BnlsniainnceiB

Rhizome Peri. „ Int.

Cotyledon Sees coat

Endo.sperni Endocarp

(See Table II)

li.

Acbene

Lf.

Lf.

Rhizome Peri. Int.

Boot. Int.

Lf. Head

Lf. Head

Lf.

Root Lf.

Seed

lY V

+ IV

VI

x(ÖR)

I

1 +

m

VI

in III

IV

+

IV V VI VI

x(Br) X (I5rY)

YI VI

II

IV II I

IV

IV

V VI

III

+ II

+ +

IV

III

Remarks

The skin brown. Yellow tissue.

YI I Brown, astringent.

Oily white tissue.

Unripe, colourless.

The cortex pxirplc Yellow If. ■VMiite If. Green If. "White If. ■White stem.

AVhile tissue.

Unri]« seed, brown wlien riiH.'.

A GENETICO-rHYSIOLOCtIC.\L STUDY ON THE F. iJiMATlDN ETC.

31

Nnmu of plnnt	Pnrt uxnmmeil	Chromogen p y	Bemarkes
Iipjxttiens Balsamiivi	Soe.1	I	IV	Unrii« seed brown wlicn riiK-, flower magenta.
ti "	•■	I	+	Unrijie see<l flower white.
Eutaoefe
Citnts Aumnihim JmiOi	Eiiil. IVri.	x(0)	VII	Pape fniit.
f.	„ Int.	x(M)	V	■\Vhite tissue.
	Fruit ' imlp '	x(0)	+
»» »• »•	Sftd	x(0)	+
C. Ateranlium amara	Kinl Peri.	x(Br)	III	Kipe fniit.
t.	Fniit • imli) '	x(0)	4-
..	See.1	x(0)	+
C. Limonwn	Einil Peri	x(Y)	III
,>	„ Int.	x(0)	II
„	Frnit • luilp '	x(0)	+	Eipe fniit.
C. GramJii	Kind. Peri.	x(OE)	IV
„	.. Int.	x(OK)	III
*»	Fniit • piilp '	x(OK)	+	Pink coloured.
„	Section (carpel)	X	IV
C. sinensis (Stivvl)	Ein.l Peri.	X	III	Ripe fniit.
„	„ Int.	X	I	AVhite tissue.
.,	Fniit • pull) '	X	^
C. nolilis var. Uiiskiu	Kind. Peri.	+	in	lüpe fniit.
tt »1 ?»	„ Int.	—	HI	White tissue.
•* '■>	Fruit, • pulp '	+	VI
Vitrtceie
VUis	(See Tables 10, 12)
Rhiimnftcete
X'gyphus v'l'.'jayis vur. ifiermis.	Frrat	I	VI	unripe, green fnut
S:vxifmgiice£e
iJiies grossiiiaria	Fntit ' skin '	in	+	unripe green fniit.
..	' fre.sli ' it seed.	III-	+
..	Lf.	1+	1+
7.'iics sp. (ciirnmt)	Fniit	ni	+	Green fruit.

1. A beautiful retluction coloiu is partially due to hesperitlin which is present in the rinp of oranges and by reduction, yields a marked reduction-colour hke tiavones. This information, the \vri er owes to the kindness of Prof. K. Shibatn.

32

I. NAGAI;

Name of plant	P.irt examined	Chromogen r F	Eemarkes
Hihes up. (currant)	Lf	1+	1 +
Oenotheraceie
Of. Lamarckkma	Li.	+	m
Cmbelliferoe
Otnanthe sioloniftra	Koot. Peri.	VI	—
»» »1	Lf. & stem.	VI-	HI
Daucus Carolit	Eoot. Peri.	-I-	—
»' )»	.. Int.	+	—
Araliacere
Aralia corJata	Lf.	-	I
Kosacene
Cydon'a itihjaris	Fniit. Peri. „ Int.	II- If	VI -f	PapülK removed which are rich in flavone.
Pseuio-cydonia sineyisis	Fnut. Peri.	I	VI
	., Int.	1+	+
Eryolotyria japonka	Fniit. 'skin'	I	III
»» II	' fresh '	VI	-
Fragar'ui chiloensis	Fniit ( receirtacle) A: .ichenes	II	VI-	Unripe, colourless receptacles.
Malus sylvestrui
" Eawles Janet "	Fniit. Peri.	I+ +	VI	I'nrii^e fruit.
	Int. (cortex of receptacle)	I+-r	+
" Jonathan "	Fruit Peri.	I+ +	VI
>.	„ Int.	I+ +	VI-
"Ben Davis"	., Peri.	I+ +	VI-
>■	.. Int.	I	+
"Smith CidcT"	,. Peri.	If +	VI-
..	„ Int.	I	+
'•Twenty Ounce"	„ Peri.	If ^f	VI
,.	.. Int.	I+ +	+
'■ Iwai "	,. Peri.	I	+
••	„ Int.	n	+
I'rmms communis	,. Peri.	III	V
>■	,. Int.	nr	+
firus serot'.na	.. Peri.	V	—
rriimis ^fmne	•<	II	+	Fully ripened fruit
"	.. Int.	II	. —
J'. lri)lo)-a (Siunomo)	•• ■>	II 1	~~ i	Ejiioarp deep rel.

A OENETIC0-rHYSI0LOOIC.\I, SITDY ON THE FORMATION ETC.

33

Nnine of plnnt	Tort exnmine«!	Chromogen	Remarks
Prunus triflora (Hnt(nnkyo)	Fruit. Peri.	11+	VI-
..	., Int.	11	4-
Ilosa rugosa	Koot & rhizome	I	I
Jl. miiUifiora	JA.	+	I
»» »«	Petals	4-	:i	■White petalH.
riiinlionncea;
huVujofera pseiKMinrtoriu	Seotl Pod	I I	\i	Unripe green seed, (lark brown when ripe.
Cytifiis scoparius	Shoot	x(0)	-1-
,.	L'ark	III	IV
> »»	■\Vootl	-r	-
rhaseolas vulgaris	(See -Table 9)
Ph. radiatus vnr. fiiirftis	(See Table 8)
Glycim soja	(See Table 20)
Pisum stiivum	See<l	1+	+	Cnri|)e seed, brown when rijie.
T'icirt Fa6n	Seed coat	* II	x^O)
»» »»	Cotyledon	-	~
J» fi	Pod	-	y
Vicia sativa	■Li.		I
»» »»	Seed	11+	x(0)
Trifolium pratense	Lf	-	n
T. repens	»»	—	V
Ebenacese
Diospyros Kaki
" Tsurano-tomo "	Fniit. Peri (epi.- & meso-carp)	II	VI	Sweet, with brown SJXltS.
,.	,. Int.	II	+	.•
" Fuyu "	„ Peri.	III	+	..
„	,. Int.	III	—
" Amn-hyaknme "	.. Peri. ,. Int.	I I	+	Sweet, no brown spots.
" Hynkiune "	,. Peri. .. Int.	II in	VI	Sweet, Iirown siKitteil.
" SUogetsii "	,. Peri.	It	VI	..
,,	,. Int.	ni+	-
" Jiro "	,, Peri.	II	VI	„

34

I. NAGAi;

Käme of plant	Part examined	Chromogeu	Kemarks
" Jiro "	Fniit. Int.	ni	—
■' Hfina-gosho "	.. Peri. .. Int.	II in	VII	Sweet, no browTi spots.
" Zcnji-mara "	.. Peri.	]i	V	Sweet, many sjx)ts.
.,	Int.	11 +	VI-
" Hiiku-Dyii "	,. Peri.	III	+	Astringent, no sjxjts.
fi ft	Int.	11+	—	„
" Wase-jisha "	.. Peri.	m	+	„
>.	.. Int.	II f	—	»»
Unnamefl	.. Peri.	1 +	II
„	Int.	If	III	„
" Yokono "	„ Peri.	III ^	VI	„
„	„ Int.	II	+	„
" Fane-nnshi "'	.. Peri.	II	VI-	„
..	.. Int.	11+	-	»t
" Mishiraclsii "	.. Peri.	11+	-	»»
	Int.	n+	+	»»
" Piiji ■■	„ ' Peri.	11	VI	1»
	„ Int.	11+	VI-	„
" Dojo-bacliiya "	.. Peri.	n	VI	»»
<• t<	.. Int.	II	+	„
" Yotsn-ya "	., Peri.	1 +	VI-
,.	• ,. Int.	1 +	+	„
Inayama ' '	„ Peri.	11 +	V	„
	„ Int.	11+	VI	»»
CoDvoIviilacen;
Ipormea Satatas	Root. Peri.	—	-	The cortex ' white'
	„ Int.	—	-	Yellow tissue.
Folanaceio
Solanum Melongena	Fniit	x(BG)	—	Sldn deep piiriJle.
S. tiiberosum		+	-
" Nemuro "	'Txihvr	+	-	Skin green
" „ Scc'illiug "	Peri.	+	-	Stem green
" White Rose "	Lf.	—	IV +	„ »
" Hayes Kidney "	.,	—	IV-	» »
Nicotiumi Tdbamim	Lf.		III-	Creeu leaf.
Datura stnimonium	„	-	-
>• »	Flower	x(0)	x(0)	White petal.

A OliNETICO-l'HY.SIOLOGICAL STUDY ON THE FORMATION ETC.

35

Nnme of pliint	Part cx(uuine<l	Chromogen r F	Rcmnrka
Ijnbiftteio
Mtmih't arvtiuiis viir.	Lf.	__	IV
pijjfrascens
Ciu-urbitnce«)
C'iritrhUa inoschtitii var. Tootvis	Fmit. Peri.	+	+	Fully riped, yellow.
.>	„ Int.	+	—
Ottaimli sotivus	.. Peri.	-	VI
..	.. Int.	—	—
Comijositero
Aniium Lappa	Koot. Peri.	—	—
,.	„ Int.	—	-
Ciirthamtis tincloritt	Lf.	x(BG)	x(OK)
	Flower bud	x(15G)	xroE)

IL Genetical Study.

1. The Mode of Inheritance of Anthoctanin and Brown Pigment in the Awn and other Parts

in ORYZA SATIVA.

In the preceding chfiptei-s, certain physiological relations in the formation of anthocyanin and the reddish brown pigments to the chromogenic substances are discuased. We shall now crjusider the genetical factor's relating to the formation of these pigments in the awn, paleas, glumes and the grain of Oryza sativa.

(a) Colom- Types of the Awn.

Among the number of cultivated varieties of Orym sativa var. idilissima and 0. fi. vai'. (jhdinosa, many ai'e awned. They may be gi'ouped under the following types with respect to the colom- of the awn.

1. Awn with anthocyanin.

a. Pm-ple.

b. Refl.

2. \.v;u without anthocyanin.

36

I. NAGAI

a. Bi'oNvu \\lieu l'ull\' lüatiu-e.

1). Faint yellow -svlieii fuUv matui-e. The brown and fcdut AeUow awns are gieen when young and the alcoholic extract of the gi'een material of the brown avm \-ields a distinct red colour by reduction but that of tlie faint yellow, only a shghtly red tinge. If oxidase is added to the exti-act, a marked brown colom- is produced in the former, while in the hxtter, practically none. It provas that the colom- of the bro^^■n awn is due to the pigment jjroduced by the oxidation of the chxomogenic substance at the end of the gi'owing period of the plant. The exti'act of the matured brown a^v^l gives also a distinct reduction colour showing that a pai-t of the clu-omogeuie substance remains without undergoing any serious change. In the matm-ed awn, the bro^\•nish yellow substance, sometimes in aggi'egates, filLs the cell, and when ti-eated with ammonia, yields a deep yellow colom-. The relative strength of the chromogeu content of certain varieties are given below.

Table 14.

CJiromogeD tontent of the extract of immature green awn of Orym saUva.

Käme of variety	Colour M'Len mature.	C'hromogen I' F	Oicitlation colour '
" Seldyama "	Brown	_	n	Brown
" Uhei "	,,	—	II-
" Kura-fusagi "	„	-	IV	„
" MeLrinsen "		—	m
" Daijo-shiro "	Faint yellow	-	VI
" Chujo "		—	V
" Koshin-den "	„	-	V	Pale yellow
" Yamato-cLikara "	„	—	VI +
" Tanpo "	„	—	V +
" Shirahige "	..	-	VI
" Nagoya-sliiro "		-	VI

Unhke the awn, the leaf does not vary ■\-er\- widol\- in the clu'omogeu content. The result of the tests made over 120 vaiieties at the middle of August, 1919, is given in Table 15.

1. Pressecl sap of tLe radisli root and hydrogen peroxide is added.

A GENETICO-rilYSIOLOGIC.Ui STUDY ON THE FORMATION ETC.

37

Table 15.

Chromogen content of the extmct of lenf in Ori/za sativa.

The colour saile wns mnde by the flnvone isolated from the leftf of

Ori/za in.stenc1 of queroetin.

Frt'iinency of vnrietieH with res]ie<;t to tlio cijlunr 'if

Colour scale		awu unii 1	ijxix oi lue J)	nieas.
	Faint yellow	Brown	Ked	Purple	Total
ni	1				1
IV	3	]		1	5
V	40	H	20	15	83
VI	18	12	3	1	34
Below VI	1				1
	03	21	23	17	124

Ac<»rding to the spech'ogi'apliic investig.ition of Y. Shibata and Kijro- TSUKI (1918)', the fibsoi-ption spectra of tho fluvone isolated fi'om the leaf of Oryza sativa conformed to these of pm-e luteoliii.

(b) Tlio Cross : Brown x Faint Yellow.

Tlie lu-eediug expeiiments made Ijy the different investigators in Japan, Imve ali-eafly shown that the hylnid between certain varieties having the brown awn and the faint yellow, gave in i^, the i-ed a-mied plant. The red coloiu' is due to anthocyauin. The segregation in I\ takes place by the ratio 9 reds, 3 browns and 4 faint yellows, otherwise the ratio is more complicated.^ Tlie phenomena are in accord mth the well known cases of the flower in Lcilhjrm in which the colourless types produce the coloured one in /'' . The wiiter obtiiined the similar results by the following crosses, "Daikkoto" x "Togo" " Kiira-fnsjigi " x " Nagoya-shiro "

The ch/xractei-s studied in tho alwve crosses are the following :

1. Shibata, Y. and Kimot4T7ki, K., Sjiectro-analysis of the Plant Pigments of Flavone Group. I. .Jour. Chem. Soc., Tokyo. 39.771, 1918. (In .Japanese)

2. Some of the results obtained by Dr. Kato are briefly summerized in Ikexo, S., Zikken- Idengakm. p. 84, 1918. .^so in Itotanic«l Abst. 2:114, Entry G79, 1919.

38

I. KAGAI :

	Stigma	Awn	Chromogen in awn <■	Paleas	Glume
^ " Daikkoto "	Colourless	Brown		BufiE	Bus
^ "Togo"		Faint yel.	VI	„	„
2'\ " Dai." X " Togo "	.,	Red	1 +	Brown	Bed
^ " Kura-fusagi "	„	Brow-n		„	,.
•^ " Kagoya-shiro "	,.	Faint ycl.	y	Buff	Buff
7'', " Kurn." x " Nago."		Red		"	Be.l

The red a^vll cliauges to bro^\u wlieu fully matui'e and becomes iuclistiu- giiisliable from the browu one at the time of harvest. The ptu-ple is luiirh more stable than the red, heiice the purple remains unchanged even at the time of harvest. The segi-egatiou observed in the i^ generation was as follows :

Table 1G.

Showing the result obtaineil in h\

Kind of cross	Awn red	A\\n brown	Awn faint yel.	Totals
" Daik." X " Togo " Pt. I	77	It	24	115
II	113	27	42	182
	TIKI	41	fir,	297
Expect. (9:3:4)	iiii; '.n;	.55. (V.»	7125
Probable errors	± 5.765	± 4.537	±.5.033
Diff. (ob.— expect)'	+ 23.04	-14.09	-8.25
"Kura."x"Nago." Pt. I	85	25	32	142
II	125	43	(W	236
	210	(!8	100	378
Expect. (9 : 3 : 4)	212.(14	7i;.85	94.50
Probable errors	±r..5()fi	±5.118	±5,ß79
Diir. fob. -expect.)	-2.C.4	-2SS	+5 50

1. One gram dried awn was extracted with 30 cc of a 45 jK^r cent alcohol. Five ce of the extract were reduced by mean.H of one cc of concentrated hydrochloric acid and the due iimomil of magnesium powder.

A GEKETICOriiYSIOLOGICAL STUDY ON THE FORMATION ETC.

39

In tlio ciiso of tho cross " Diiikkuto " x " Togo " reils uuil liruwus were oftou difficult to distiugiiisb jiccnratcly owiug to the fcviutness of the refl pigiueut aud the rapidity iu chauge t(i tho hrownish colour. A considerable de\iatiou fi'om the expectation thus aix)se, though the ol)seiTed nunibere are fairl}- close to the approximation to a 9:3:4 ratio.

Thh-ty six browns and twelve faint j-ellows were raised in the next year. If a 9:3:4 ratio is accepted, we shotild expect among browns wliich include reds and lirowns, two gixjups of families in the I'\ generation. One of them should throw again red and brown, and the other should not throw red. The ratio of two such families should bo 3 : 1. "We olitnincd :

	Xo. of fiimilifs which threw reds	Xo. of fnmilies whk-h threw no reils	Total
Observed EsiJected	32 i 27 9	36 36
DiScnnce	+ 5.0+1.701	-.5.0

Among the red throwing famihes, tho tullowing famihes are expected :

Type of families	begregati.jii
1.	I?cil constant occurring once in every nine.
2.	Beds, browns and faint yellows in 9:3:4 ratio occurring four
	times in every nine.
3.	Eeds and faint yellows in 3 : 1 ratio occurring two in every nine.
t.	Reds and browns in 3 : 1 ratio occurring twice in every runs.

Among the famihes which tlii-ow no i-eds, the following famiUes are expected :

Type of families

Segregation

&• Brown constant, occurring once in every three.

6, Browns and faint yellowish 3 : 1 ratio, occurring twice in every

three.

We found :

40

I. NAGAI;

Type of families	No. of families observed expected	Difference (ob.— exp.)
1.	3	3	0
2.	7	6	+ 1
3.	10	6	+4
4.	12	1-i	0
5.	2	3	-1
6.	O	6	-4
:u'.	3(i

Tlie iuiut yellows are exjoectal to he coustiint Twelve families raised \\ere all constant. The actual uumbei-s are given in Table 31.

In tlie case of the cross " Kiu'afusagi " x " Nagoya-sbii-o ", the red coloiu- of the awn was more distinct than the former case, hence the agreement of the expected to the observed numbers was close. In the i^, %vas obtained:

Type of families	Xo. of families obseired expected	Difference (ob.— exp./
1.	8	7.583	+ 0.417
2.	31	30.332	+ 0.6G8
3.	13	15.1C6	-2.166
4.	18	15.1f.6	+2.834
5.	6	7.583	-1.583
6.	15	15.16G	-0.166
	91	90.996

Thii'ty two faint j-eUow gave all constant families. See Table 32.

(c) The Cross : Red x Pmple.

The relation of the colour of the stigma, awn, palea, glumes and the leaf-sheath was studied by the cross " Hanbim-neuto " x " Grem-okumochi ". The charactei'S involved in tliis cross wliich were subjected to investigation %\cre as f )llc )ws :

Avm

■ Hanlnin."

Red

'Genroku.'

Pnrple

F, '• Hanbim." x " Genroku.'

Purple

A GENETICO-rHY.SIOLOGICAL STUDY ON THE FORMATION ETC.

41

	" Hftnbnn."	'• Gemokti."	F, "Hftnbun."x"Gcncokn."
Stignift	Colourless	Eeil to imriile	Ked to ptirple
Pftlea	Brown	rnrjile localized in streiiks from the tip to down- ward on yellow ground colour	Self purple on brown ground colour
filume	Eid	Purple	Purple
Lejif -sheath	Green	(Jreen with imrpk- stripes	Green with purple stripes

Tliiis we see that the doiniuciut foloni-s are ; iu the awu pm-ple over reel, in tlie stigma reel over coloiu'less, and iu the leaf-sheath pm-jJe stiipecl over non-striped. A new type of palea was found iu the i^ plant, namely the self piu'ple on the bi-owu gi"onnd coloin- (see Plate I). Tlie brown gix)und-colour is dominant over yellow, and the self piii^jlc is domiuant over the above two, lieing epistatic to browu. A («mpletc liukage is found between the self pur]^)le !iud browu giwuid coloiu'. The loc-aJized pm-ple is always uon browu with respect to the gi'ouud eolom* in the paleas. Iu the jP. geueratiou, we obtained the following results.

Table 17.

Showing the result obtained in F.,.

	Stigma	coloured	Stigma	colourless	Total
	Pt. I.	Pt. II.	Pt. I.	Pt. II.
.\wn purple	ö'J	49	—	—	108
.\wn red	—	—	12	22	34
Total	Ti'J	40	12	22	142
Paleas, self purple (deep)	i	0	—		10
(med'um)	18	16	—	—	34
(pale)	U	IC.	—	—	30
„ purple ocalized (prominent)	•i	3	—		7
„ purple (medium)	16	8	—	—	24
„ brown		—	8	15	23

42

I. NAGAI:

	Stigma	coloured	Stigm.'X	colourless	Total
Palens, yellow	1?	—	4	7	12
Total	571	49	12	22	140
Glume, purple red „ imdetermined	59	49	8 4	22	108 30 4
Total	','.)	49	12	22	142

By Slimming np tlie above figures, we obtaiu

	Stigma coloiued & Icaf- sheath striped	Stigma colourless & leaf- sheath non strii>ed
Observed Expected (3:1)	108. 106.5	34 35.5
T>iS. (ob.— exp.)	+1.5	-1.5

With respect ti) the oi)Ionr (if the palea :

	Self purple on brown	L,x-alized purple ^^^^.^ bellow on yello»'
Observed Expect. (9:3:3:1)	74 78.75	31 26.25	23 12 2G.25 8.75
Difif. (ob.-exp) Probable errors	-4.75 ±3.959	+ 4.75 ±3.115	3.25 + 3.25 ±3.115 ±1.9:12

The above ratios were confirmed l>y the result obtained in F^. The actual numbers are given in Table 33.

The coloured stigma, pm-ple awn and glume, self jMu^jle palea, and stiiped leaf-sheath are inherited together. According to Hector (1916),- the Indian varieties stuched at Ducca, may be classified as follows ;

(1) Leaf-.sheaths, apicidus of glumes, and stigma colom-ed.

(2) Leaf-sheatlis and apiculus of glumes coloured, but stigma colourless.

1. Two plants died.

2. Hectob, G. p.. Observations on the Inheritance of .\nthocyan Pigment in Paddy Varie- ties, llemoits Depart. .\gri. India. Bot. Series. 8 : Xo. 2, 89, lOlG.

A GENETICO-PHYSIOLXGICAL STi'DY ON THE FilRMATION ETC.

43

(H) Apiciilus (if j^liuues oolonriMl, Imt leul'-slieatlis ci Joiirless.

(4) Apiciilas of glumes ouly colonitifl. He douljts of classes .3 and 4 roiilly exist. Chtss 1 is the commonest gi-oup. Coutrarv to ludiau varieties, the cxalom'less stigiim with gi'cou leaf-sheath is the most common type fuimd in JjiiMiuese varieties. He observal further, thut the c<jl(>ur in the leaf-.sheatlis and apiculus is dne to a colour factor R acting on a du'omogeu C, and that the jmrple colour of the stigma is duo to a farther factor P not present in the leaf-sheath and apiculus, and that the simultiineous presence of all three fcwtoi-s liCP is necessai-y for the protluction of any sort in the stigma. Paenell, Ranoasw^vmi, Agyaxgai;, and Raxi.vh (1917)' have shown that in certain Indian varieties, the pui^ile lining of the iuternode is coupled with tlie purple glumes, the purple stigma witli the praple axil (purple-coloui'iug of the epidermis on the inside of the sheath) but the puiplo lining of the interuode and the purple glume do not co-exist with the piu'ple stigma and the piurple axil.

Among the Japanese varieties studied by the writer, most of those which have a colom-ed stigma have a puiple awn and leaf-sheaths striped, and only in certain varieties, they are separated. The distribution of authocjanin in Japanese varieties is sho'(\ n in the following table.

Table 18.

Distribution of antliocynniu in the \arieties of .Tnianese rice.

Name of typical Tariety	Stigma	Awn	Palea (tip)	Palea	Glume	Lf. sheath	Lf. bLade
" Murasaki "	Coloured	Purple	Purple	Purple	Ptirple	Puriile	Purple
" Totachi-wase "	n	••	■•	Red	Eed	Purple striped	Eed striped
" Haguro "		"	"	Purple	Purple	Ked strip.	—
" Edowase "		.,	)»		—	„	—
" liozu-karasu "		Awnless	„	—	,.	—	—
" I'yejini-shirazu "		Purple	Keddish brown	—		—	—
" Gtnrokii-mochi "			—	Purple	Purple	Purple striped	—

1. Pabxell, p. E.. Eangaswajii, Agtangae. G. N, and Rvniah. K., The Inheritance of Characters in Kice I. Memoirs Depart. Agri. India. Bot. Series. 9 : 75, 1917.

44

I. NAG.U:

Name of typical Variety	Stigma	Awn	Palea (tiPj	Palea	Glume	Lf. sheath	Lf. blade
" Choja-bozu "	Coloured	Awnless	—	—	—	Purple striped	—
•' ]iiinta-bozii "	„	„	—	—	—	—	—
" Isejiro "	„	Purple	Purple	—	Piuple	—	—
" Gorobei "	Colourless	Awnless	Ked	—	Red*	—	—
" Hozoroi "	„	,.	.,	—	„	— .	—
" Homara "	.,	Red		—	„	—	—
•' Asaterashi "	-	"	"	—	"		—

The puiple iiud reel a^\n is greeu wlieu the panicle is iu the leaf-sheath. The red colour begins to develop at fii'st at the tip and the base of tha awn, a day or two after the panicle has appaared from the leaf-sheath (see Plate I). The colour gi'adually extends to the entii'e portion and at the same time increases in intensity. The pm'j^le awn is red iu the beginning, but rapidly inteusifies in ce)lour and becomes deep piu'ple. In the red awn, on the other hand, the red pigment remains unchanged and sooner or later, it is decomposed. A similar change is obssrved in the glume.

The development of the pigment in the awn is depaudent on the illumi- nation. When the panicle .is enclosed in the papar bag to ensm-e self pallinji- tion, the pigment develops only slightly. In th^ stigma, purjjle anthocyanin is ah-eady present even when the p.anicle is still in the leaf sheath. Tims the development of anthocyanin is saen to have difierent physiological requii'e- ments even iu ths difi(?rent parts of the same floral organs.

In the cells of the colourless stigma, flavone c:m be detected by ti'eating them with ammonia which jield a deep yellow colom-. It is quite probable, however, that there may be a colourless stigma having no flavone (chromogen). Hector (1916)' has shown that the colour of stigma in certain cases, is due to more than three factoi-s.

Tlie anthocj-anln pigment in tlie palea is coufinal to the epidermis, and the brown pigment to the underlying tissue. The brown pigment is practically insoluble in strong alcohol, but slightly iu a weak solution. Fully matural

* The colour indistinct, sometimes obscure. 1. Hectob, Ci. P., Loc. cit.

A OENETICO-PHYSIOLOOICAIi KTITJY 0\ THE FoItMATHiS VTC.

45

I'i(i>\ii piileas were exti-acted \\itli alculml fur n luug time, ami the extract was tested for i-lin)iui)p_Mi. Pi-actü-ally iki rod coluur was fimud 1>y reducing uor Ii_v healing' witli liydnicliluric ncid.

Aiuoug the Fi plants certain auo- CL a! lualies which nre worthy of mention,

were found.

Two spikelets born ou a panicle of a plant which bore the red awn and _^ello\\- paleas, jwssesseil the browu in- ferior palea.

Two spikelets born ou a panicle of a pLiut lia\iug the spikelets which liore a purple a^^■n, paleas and glumes were sectori:üly pigmented with purple and yellow (see text fig. 2.)

T^\o gi'ains were found in Ji single spikelet in one of the spikelets of a normal plant. The spikelet of Oryza scdiva normally beai-s only one gi"ain.

In one plant, both paleas, inferior

aud sujxjrior were foiuid to bear awn.

The anomalous awns were shoi-t and

red coloured. Normally, the inferior

palea only possess a long awn.

In one of the spikelets born on a panicle of a plant, the lodicules were

found to be modifietl to small ivilea-like appeudtiges which were nan'ow and

shai^jly pointed.

Text fig. i. Showing the nnomaloiLs sec- torially pigmented paleas. Bom on the panicle of which spikelets were self purple, (t, a', same spikelet ; h, V, same spikelet seen from the different sides. p, piirjilc area.

(d) The llelation Between the Colour of the Grain and the Paleas.

It is of interest to obsene the diflerence in the kind of clu-omogeuic sul>- stiuico of the brown pigments in the awn and the gi-ain. The coloiu- of the gi-ain of most of the Jajmnese varieties is pale buft" but in hw, reddish brown.

4G

I. XAGAI ;

The clu-omogenic substance eau be detected in the extract of the milk ri^ie, green grains of the redtlish brown sort but not in that of the pale buif.

Table 19.

Chromogen content in the extract of the milk-ripe grain of OtxJki sntiva >.

Name of variety	Colour of fiilly matured grain	Chromogen P F	Colour of extract with peroxidase with KOH (boiled)
"Dhei"	Pale buff	-	-	—	Yellow
" Kinoshitamochi "	..	-I-	-	-	.,
" Kameno-o "	,.	-	-	—	„
'• Togo "			-	-	„
" Seki-yama "	„	-	-	-	„
•' Yamato-cnikara "	„	-	-	-	„
•• Akamoro "	Eeddish brown	II	—	Orange brown	Deep blue to re<F
" Haguro "		"	~

As we have already seen the clu'oiuogen in the awn belongs to the group of the chi-omogenic substance F, and in the gi'ain, to the chroniogenic sub- stance P.

According to Kondo (1917),' the redtlish brown pigment of the gi-ain is confined to a single cell layer of the integument of the seed coat. In the dark colom-ed gi'ain of the foreign sort, the -«Titer observed that the pigment ivas not confined to the seed coat. An anthocyanin-hke pigment was foimd in the jiericarp, namely tjie layer above the tube cells and even in the grain of tlie brown sort of Japanese rice, a yellowish pigment was found in the cell of the pericarp.

Paenell et .Al^ have shown that in Indian rice, the red, grey brown, and

1. Three grams of fresh material were extracted with 15 cc of a 40 ]X?r cent alcohol.

2. The extract was boiled with catistic potash. A deep blue colour like that produced by the solution of myricetin was found. On cooling, the blue calour changed to red. It may in- cidentally be mentioned that most of the flavones give a deep yellow colour when treated with caustic jiotash, but myricetin gives a deep blue instead of yellow.

3. KosDo, M., Untersuchung neber ilic Dicke der Beiskleischicht. Bericht, d. Ohara Inst. f. Landwirt. Forsch. 1 : 219, 1917.

4. See also lÜKK.twA, S., On the Classification of Cultivated Eice. Jour. Coll. .Vgri. Imjier. I'niv. Tokyo. 3:11. 1912, Pakxell et al. Loc. cit.

A GENETICO-PHYSIOLOGICAL STUDY ON THE FORM.VTION ETC.

47

white graius scgregattxl l>y tlio latio 9 : !i : 4 iu the desceudauts of certsiiii mitiu-al hybrids, and the red grain plants were purple-pigmentetl whereas the gi'ey-brown plants were nnpigmented. A similar «ise was found in the cix)ss tetween two Jajmuese varieties, " Otsubu " x " Hagiu-o ". Tlie grain of tl:e former wiis psde biiH' ("white") and the latter was roddisli brown. Tn tlie F. plaut.s the following result was obtained.

T.VBLE 20.

■	Showing the result obtainecl in •' Otsnbu " X " Haguio "	f».
Culuur vt	Culour of grnins
	re<ltlish yillow brown brown	jmle buff (white)	Total
Awn Paleas
Purple Self piirplc Purple Loa piuple Brow-n Brown Yellow	119 33	33 12	36 12 11 4	155 45 44 16
	152	45	C3	260
Kxpect (9:3:4) DjfE. (ob. — exp.) Probable errors	14G.25 + 5.75 ±5.375	48.75 -3.75 ±4.185	C.5.00 -2.00 ±1.007

The yellowish l)rowu gi-ains were not fovmd iu the purple awn plants. The ratios of coloured to white grains in each awn type showed a normal 3 : 1 ratio (see Table 20), so it is certain that a Hukage relation exists between the genes for the pm-ple awn and for the reddish brown grain. With respect to the colour of the awn, pui-ple and brown showed a 3 : 1 ratio, and iu the palea, self pui-ple, locahzetl piu*ple, brown and yeUow segi'egated by the ratio 9:3:3:1 (155 : 45 : 44 : 16 obseiwed against 146.25 : 48.75 : 48.75 : 16.25 ex- pected) aa in the case with the cross " Hanbuunento " X " Gem-okii-mochi ", already reported in the section (c). These two crosses differ with respect to the colour of paleas, only from the gene gi'oups which enter from the parental plants, but the manner of segi-egation in the i^ generation is the same, thus :

48

I. KAGAI:

Kind of hybrid	Self purple	Loc. jHirple	Brown	Yellow
" Hanbun." x " Genrokii " " Otsubii." X " Haguro." F.J " Hanbun." X " Genroku." Kj " Otsubu." X " Haguio."	9	3 3	3 3	1 1

The sec-tiou of the reclchsh bruwu gi'ain showetl that the pigiueut was chiefly coufiued to the siDgle cell layer iu the testa as KoNDO obseiTetl, though iu the ]^)ericarp, a yellowish pigment also occuiTecl. Iu the yellowish browu grain, the pigment occuiTed chiefly iu the pericaip and the testa was slightly IJigmeuted. We are deaHug, therefore, ^nth the pericarp and the seed-coat colour iu two ty^jes of the grains, but iu the fully ripe grain, the two parts i\xe hardly- distinguishable unlike the beau iu wliich they are diflereutiated into the jxid and the seed c-oat.

2. The Mode of Lsheeitanxe of Anthocyaxiis" and Brows Pigment in the Seed Coat of Glycixe soja

Notliing jDerhaps excels the seed coat of the legumes iu diveisity of the colour cLiracters exhibited by the plant except the flowei-s of some omameutal plants. As -we have akeady seen, the seed coat of the coloured beau of the legumes is rich in the chroniogeuic substance previous to pigmentation. Some data on the genetical behavior of the colour charactei-s in the seed coat of soy liean {Glycine Soja) are discussed in tliis chapter.

(a) Colour Tj-jjes of the Seed Coat.

The diflereut varieties of Japanese soy beans may be classified under the following t>-j)es with respect to the colour of the seed coat. I. Self-colour t^vjie.

1. Black (deep piu-ple).

2. Retldish lirowu.

3. Brown with or without a gi'eeu tinge including diftorent sliades

of brown.

4. Buflü

A OKXETrCO-rHYsroLOrilCAT, MTrnV ox the FORMATrriN ETC. 49

5. Greeu. (j. Yellow. IT. P!U-ti-cül(jur type.

1. Bliu-k mottled, the ground colour brnwu with or without a gi'een

tiuge.

2. Black p:itch arouud the liiluiu. The gi'ouud cok>ur is either

yellow or greeu.

3. Dark browu patch around the hilutu, the gi'ound colour is either

green or yellow.

4. Blue tiuged around the liilum. The gi'ouud colour either greeu

or yellow. The margin of the Muo tiuge is not so distinct as

in 2 and 3. The chromogen can be detected in the imiuatm-e gi'eeu seed of all the coloured tj-pos, except the gi-een and yellow. Tli3 chromogenic substance P is abundant but the chromogenic substance F is very scarce or absent. In the green and yeUow, both are nearly absent. Heuce the different types cau be distinguished into two gi-oups vdih. resi^ect to the clu'omogen. One includes those ■which give a marked chromogen reaction -when the seed is still gi-eeu and the other includes those which give only a shght or no reaction. In this regard, the gi-een and yellow c-oiTespoud with the wliite tv^pe of the common garden Ijeau and Adzuld-bean (see Tables 8, 9).

Since the reaction of the chromogenic substance F is feeble in the seed just before the formation of the pigment, the latter must be formed fi-om the duromogenic substance P which is present. In the leaf, however, the chromogenic substance F occurs in a considerable amount and can readily be isolated as yellow ciy-stals. The following table will give a general idea of the distribution of the chromogeus in the seed coat and the leaf of different varieties of soy-l>eans.

50

I. XAGAI ;

T.iBLE 21.

C'hroinogeii content in the extract of tinripe, green seed, nncl leaf of Glycine soja.^

Name of variety.	Colour of set-d when mature.	Ch I'	romogt-n F	in seed oxidation colour-	Chromogen in P F
" Kxirodaidzn-ko "'	Solid black	III	—	RBr.	1	m
" GoisM "	"	III	-	EBr	-	ni+
"Goishi" (flower purple)		III	-			iiif
" Nedznmi-meta "	Br<5WTi	IV	-	OBr	- ,	nx-
" Akakznka "	„	11	+	..	-	IV
•• Cha "	„	II	-	.,	-	lUf-
" Haiiro "	„	HI	+		-	in
" Beni-iro-ilaidzti "	Eed brown	III	-		-	ni4-
" Akanedzninime " " Madara "' " Juseita "	1 Black mottled 1 brown Black patched	III lit IV	+	^	-	m UI + ni
" Kiira-kake "		UI	-	„	-	lU
" Achumuri " " Tanishi " " Tora-mame "	Sohd black fBliie tingal \ yeUow Solid black	in V IV	VI	Yßr OBr	-	ni nn- ni-F
" Goku-ao "	Green fgreenj'	-	-	—	-	III
" Aobisbi "	.1	VI	-	—	-	IV
" Uma-daidz« "	„ „	-	-	—	-	III
Goyo "	..	-	-	—	-	ni
"Ao"	,.	-	-	—	-	III
'• Toyo-naga "	Green (yellow)	-	-	—	-	m-
" Dateao "	») *i	-	-	—	-	III
" Yoshioka "	<> <•	-	-	—	-	IV
" Shakujo "	YeUow (yeUow)	-		—	-	ni+
" Shiro sota "	..	—		—	-	ni
" Omejiro "	..	-	-	—	-	III
•• Yukinoshita "	*• V	—	-	—	-	III
'■ Chogetsu "	.>	-	-	—	-	III
" Kimusume "	,.	-	-	—	-	rv'

1. The extraction was miule w ith 5 oc of a 10 per cent, alcohol to each gram of fresh beans and with 10 cc for leaf.

2. Tressed juice of potato tuber mil hydrogen i>eroxide mldeil. The designations of colours are the same as those use«! in Table 13.

3. The colour of cotyledon.

A OENETico-rHYsroLooicAL sTi'ny ON' Tirr. r )n>r.vTioN etc.

51

Uame of variety	Colonr of seed when nuitiue	Chromogen P F	in seed oxidation colour	Chromogen in If. P F
" Sennnri "	., .,	—	—		_	Ill
" Knribntnkiyn "		vr	-		—	ni-
•' Sliiro nixlzuiui "			-	—	—	ni
'• BMkni-shiro "		-	-	—	—	IV
" Slionm-wnse "		-	-	—	—	in
' Shiro-lmchikolfii "			-	—	—	IV
•' Abiiru niixnic "		-	—	—	—	in

TliJ piu'ple and IJiie pigmoute are iintliocyanins. Botli anthocyaniu ami brown pigments are coufiued to the epidermis existing in the same cell. Wlieii a common black bean ls boiled with an allailine solution, the colour of the solution becomes blue to gi-ceu at first, but soon changes to deep wine red owing to the extraction of the brown pigment in alkali.

Tlie poi-oxidase wjis tested in a number of varieties at the stages previous to pigmentation of the seed coat. Tlie reaction was very distinct in the seed coat of all the coloured types as well as in the yellow and gi-eeu ones, but the direct oxidase reaction failed by alplia uaphtol, l^nzidine, myiicetin and guiiiacum. It was ofk-n noted when alpha naphtfjl was apj)lied, the reaction was veiy faint in the epidermis even in the presence of hyctogen peroxide. Tliere could be fimnd no definite diiferenca with respect to the peroxidase reacHou exliibited by the seed coat? of beans which contain the chromogenic substance and those lacking it;

In the black patched Ijean, however, the part where the anthocyanin coloin- will ajipear in a patch, jielded a deep pm-ple or bro^ni colom- by immersing the whole bean which was previously ti-eated with alcohoUc solution of benzidiue or nnnicetin, in a dilute solution of hydrogen peroxide. Tlie remaining part where anthocyaniu will injt be formed at all, yielded less distinct colour in the case of benzidine and nearly failed in the case of mpicetin. The chromogenic substance was detected in the tissue of the n(ju- IjLick part. Tlie alcoholic extract of the portion gave a slight red colour by heating with hydnx;liloric acid.

The above oljseiTation seems to point out the fact that at the time of pigment formation, the black patched portion of the seed coat contains more

52 I. NAGAI :

active \5er0xidase system tliau the rest of the pcntiou where the developmeut of the anthocyauin pigmeut is inhibited, iu spite that the chi'omogeuic substance cau be detected iu a shght extent.

The above obsei-vation agrees with the findings of Keeble aud Aemstrong (1912j' who have shown the paralleHsra extistiug iu the distribution of anthocyauin and jieroxidase iu the corolla of PrivivJa sinensis and in others.

When the seed coat is still gi-cen just previous to the formatiou of anthocyaniu, a distinct red coloiu* develops at the poiiion of black patch b^' treating the beau with hydi'ocliloric acid iu cold. It shows that the authocyaiiiu pigmeut is ah"eady present iu a colourless state.

Oxidation seemed to accelerate the development of anthocyauin in the seed coat of the soy bean. Sh'ghtly coloured beans rapidly dee^jeued iu colour if the pod was opened and exposed to ah'. Injmy also accelerated the development of the pigment. Tlie portion near the injmy tecame pmple at first ou exposm'e to air. Conversely the exclusion of air retarded the develop- ment of the pigment. The sHghtly red-coloui-ed beans ceased to develop the pra^Je pigmeut when kept in a closed chamber iu i\hich the air was i-eplaced b}- hydi'ogen gas. Similar phenomena were observed iu rii^eniug gi-ajies. In both cases, light had no influence.

(b) The Cross : Blue Tinged Yellow x Brown aud the Kec-iprocal.

The colour of the seed coat of the parental plants are the following:

" Tauislu." Yellow with a tinge of blue colour ^\hicli is most prominent arouud the hilum, fading gi-adually further beyond.

" Haiiro." Browu with a tinge of gi-ejish gi'een cok)ur.

The i^ seeds produced ou the T\ plant were gi'eeu ^\•ith a pale blue tiuge as in " Tauishi " (see Plate I). Iu F, solid blacks, aud non-tiuged gi-eeus aud yellows were found, together with the parental t^^ie, by the following uumbei"s.

1. Keeble, F. ami Akmstbono, E. F., The Eole o£ Oxyiliuses in the Foriuntion of the .\ntho- cyim rigmcnt of Plants. Jour. Gent. 2 : 277, 1912.

2. Keeble, F. and Ai;meti;ong, E. F., Loc. cit.

A GENETICO-rnYSIOLOGIC-VL STUDY ON THK FORMATION ETC.

Table 22.

Showing tlic results obtiiinul iu Fj. "Tanishi" x "niiiiro."

	l>]iu'-liu'_;e(l	^■•n tiiigeil	black	Brown
	Green	YelUow	Oreen	Yellow
Ob. Ex,,.	19 17.30	6 5.77	C 5.77	0 1.92	7 7.G9	3 2.56	11 41
Diff.	+ 1.70	-f (1.23	+ 0.23	-1.92	+ 11.09	^ fl.ti

• Iliiiiro " X " Timislji.'

	lilue-tiugeil	Xon-tingol	BLick	Brown	Total
	Green Yellow	Green Yellow
Ob. Exp.	23 20.25	13 C..75	4 G.75	2 2 25	3 9.00	3 3.09	48 4.S
Diflf.	+ 2.75	+ C.2->	-2 2.-)	-U.25	-c.no	0

Tlie calculated ratio is 27 : 'J : Ü : 3 : 12 : 4. A relative small uumber in the I'] plants luade the determination of the exact ratio somewhat uncertain but by the F^ plants, tlie aljove ratio was confirmed.

T.\BLE 23.

Showing the segregation of the oftspring of the

heterozygous blue-tinged green Fi plants of the cross.

■■ Tanishi " x " •' Haiiro."

Fj family no.	Bhie-tinged	Mon-tinged	Black	Brown	Total
	Green Yellow	Green	Y^ellow
•4	18	5	4	3	5	3	38
15	Ifi	0	2	1	9	3	37
30	32	H	3	2	11	2	58
31	30	G	19	2	20	7	Si
Total	90	25	28	8	45	15	217
Exjiect.	91.55	30.52	30.52	10.17	40.09	13.56
I'rohable errors	±4.900	±3.454	±3.454	±2.111	±3.81G	±2.405
Piff. (ob.-exp.)	+ 4.45	-5.52	— 2 r>2	-2.17	+ 1.31	+ 1.44

54

1. NAOAI :

The rest of the F^ £imilies agreed ^\Uli llie expectation except tbi-ee which were apparently contaminated by accidental crossing ; hence they were dis- carded. The actual numbers are given in Table 34. See also Table 29.

(c). Tlie Cross: Buff x Black.

The colour of the seed coat of a variety " "WiU'e-mame " is buff, and that of a variety " Achummi " is solid black with pacuHar mesh-like, white marldngs Mhich are due to the small breaks of the epidermal tissue exposing the colourless, underljiug tissue of the seed coat (see Plate I). Tha F^ seeds born on the F^ plant of the hybrid between them were self black. In F^ seh IJacks, imperfect blacks, and browns were found ■v^ith the fjllowiug number.^.

Taule 24.

Showing the result obtainerl in the Fj generation of the cross " W.-iremame " x " Achumiiri."

	Self black	Imperfect black	Brown	Bluä	Total
Ob- Exp.	33 27	9 0	6 9	0 3	48 48
Diil.	+ 6.0	0	-3.0	-3.0

The expected ratio is 9:3: 3 : 1. Tlu buff wtis not found in tha F^ plauts owing to the small number of individuals. In F,,, families derived fi'om the heterozygoits self blacks, bufls were found in the expected ratio.

Table 25.

Shijwing the result obtained in Kj families deriwd from the hctei'ozygoii.s self black Fi plants.

F-, family no.	Self black	Imperfect black	Brown	Buff	Total
3	21	9	2	1	33
4	31 •	4	Pi	4	4!)
C	4ß	IH	I'J	4	87

A GENETICO-rHYSIOLOGICAL SITDY ON THE FORMATION ETC.

F.J fiimily no.	SeU binck	Imperfect black	Urown i 1	UulT	Totnl
	10	4	1	1	16
12	12	4	1	5	£2
18	2G	8	11'	'	4'i
19	3.'>	13	i:t	r,	r,7
2;}	21	7	.1	1	11
25	23	3	1	3	42
56	24	10	11	:i	5C
27	26	10	3	.1	42
33	40	11	It	i	70
Total	31.5	115 1	98	43	571
Expected	322.18	107.06	107.06 ;	35.69
Probable errors	±7.481	±e.291	±6.291 1	±3.962
Dirt, (ub.-exp.)	-0.18	+ 7.91	— \)VI\	a-7.:U

What is designated as imperfect hlack is a new type. It dlflers from the self bkck by the incomplete development of the authocyaniu pigment, resiiltiug in the brown gimind-colour being made visible (see Plate I). The gi'oimd colour of the imperfect black appeared to be brown instead of Imff, 1)ut in the F^ generation, they threw only imperfect black and l)itil' l>nt no brown. It seems therefore that the buif colour was deepened by the action of the gene which develops the anthocyanin in the same epidermal cell. The l^resent instance resembles that of the palea coloiu- observed in the cross " Hanbun-nento " x " Grenroku-mochi " in which the self jjiu-ple was completely linked with the brown ground-colour and the locahzed puiple excluded the latter. In Iwth cases, the full development of anthocyanin is in some way associated with the brown pigment.

56

I. NAG.U:

Table 26.

Showing the segregation of the offsprings of imperfect black F2 plants.

F,	family ns.	blaok imperfect	Brown	Bviff	Totals
	17	26	—	—	26
	35	27	—	—	27
	31	14	—	3	17
	3G	3J	—	9	44
	38	2G	—	5	31
	39	39	—	15	54
	40	4	—	2	6
	41	5	—	1	6

iTti

35

■ill

An approximation to a cliliybrid ratio was confirmed by forty sis families raised in the F^ generation. Tlie details are given iu Tables 30 and 35.

The biifl' is found to be the most recessive character to any other one in the colour characters so far studied. Since the chi'omogenic substance can be detected in the uuiipe, gi'een seed of buife, browns, and blacks, but is nearly absent in A-eUows and greens, and further that yellows and gi-eens are dominant over the former coloiu-s, it must be concluded that an inhibitor for the development of the pigment is present in the green and yellow.

A mention may be made regarding the brown character. We cau dis- tinguish several browns which differ mora or less in hue and shade but exact discrimination is very difficult. If it is made among the segregates, the lighter shade appears to be domniant over the deeper one. When the distinctly reddish brown such as we see in the seed coat of a variety " Aka-nedzumimo " is crossed with the brown like that we have ah-eady dealt with (" Haiiro "), the reddish brown behaves as a single recessive to brown. The F, seed of the cross " Aka-nedzumime " x " Haiii-o " was brown like that of the ftither which is brown with a gi-eyish gi"een tiuge. In K, reddish l)rowu and brown, regardless of the gi-ean tinge, segi-egated in the following number :

A GENEXICO-PHYSIOLOfilCVL STl'DY ON THE FORM.VTTON ETC.

57

	lleiliUsh brown	linjwn	Total	^
Observed Expected	C!) 6r>	19 22	88 88
Diff. i'i'b.-i-x]!. 1	+ 3.0	-3.0

The liyl)ricl between " Kari-matue " and " Akadzuka " in which the yellow and reddish brown wo.e crossed, produced the yellow in tlie I<\ plant, and in F., tli? following segregation was observed.

Table 27.

Showing the result obtiiinoil in the F< of tlie cross " Kaii-mame " x ■' .ika.Tziika."

Phmt A'o.	Green	Wllow	Black	,, Eeddish ^""° brown	Totiil
I ir	57 21	29 11	14 7	6	4 4	110 43
	78	40	21	6	8	153
Expect. Probable error Diff. (ob.-exp.)	86.062 ±4.139 -8.0G2	28.687 ±3 236 + 11.313	28.687 ±3.256 -7.687	7.172 ±1.763 -1.172	2.391 ±1.014 + 5.609

The expected ratio is .3tj : 12 : 12 : 3 : 1. Here again the deeper brown is shown to be recessive to lighter brown. A more detailed analysis of the brown characters is under waj', and the result will be reported a^ data become available.

Speaking in general, tlie inhibited colour of the seed coat can be guessed by the colour <>f tho hilmn and that of the naiTow ring around it. In the present communication, the colour of the hihim and tha ring is not considered.

.3. Ax Interpretatiox of the Kesults. It is a well known fact that there are at least two groups of genes

68 I. NAGAI :

present in relation to the formfition of authoejanin pigments in plants. One of Wiem includes those which are known as the chromogen factors, and the other includes those which are complementarv to the former. A complete system provided by the union of these genes produces the plant in which tho formation of anthocvanin pigment is realized. To designate those genes O is often used for the chromogen, and R for the complementary one. The most simple case is the counterpart of two genes C and i?. We may denote those genes which are related to the formation of the clu-omogenic substance in plants by ' chi-omogeus ' ( C) and those which are related to the formation of any biochemical agency, by means of ivhicli the chromogenic substance is conveiied to a coloured anthocyauin or brown pigment, by ' chromopheleins ' (J?, 0, etc.). According to the view put forward by I\liss Wheld.\le, the chromogen factors in the flower of Antivrldnvm are related to the formation of certain flavone glucosides (glucoside of apigenin and luteoUn) and the chromopheleins are relatid to certain oxidizing agencies probably the peroxidases.

It must clearly be understood that thi phenomena of inheritance and development are of different kinds, and the data of the latter should not be confused in interpreting the genetical data. The factor is such an entity of the organism that by its means certain gi'oups of biochemical reactions are set jree to build up the character which is the phenotypic expression of the gene. The biochemical reactions and their produces alone ai-e dealt with as physiological and developmental data. Sjme of them can well be regarded as the clue to the difference iu the genetical units, bu!: these phenomena themselves are nothing to do with those of inheritance. A well marked, different biochemical phenomena may not necessarily correspond with the difference iu their genetical potency.

It is an impossible task to know all the biochemical changes which are governed 1 ly a given gene ; all we can attempt, if at all, is to find certjiiu coiTelations between the known biochemical facts and the genetical data, by which the chief fimctiou of the gene may be inferred. Such an attempt may be useless or may fall sliort of the aim. But, as the «Titer believes, genetics aims to discover not only the laws of the mechanism of distribution of hereditary luoits, but aho the iiiiLs between the gene and the actual

OEKETICO-rHYSIOLOGICAL STUDY ON THK FflRM.VTIoX ETi'. 59

liiixliciiiicjil 111- physiological processes in tlio somatic cells that are set fi-ee ]>y Hie con-cspoudiug gonetical make-up.

In the case of the coloiir cluivacters in the awn of Oii/za mtiva, we are api>:irently dealing with instances similar to those that were observed by Miss WnKLD.OiE in Aidirrldnum and Bateson in Lcdlajrus. Suppose a pair of genes C and c are concei-ned. The gane C produces the chrt)ü:ogeuic substance in the bi-own awn to such an amount that it can readly lie detected in the extract aiid by c, the production of the same substance is as much as ten to twentj- times less tluiu that produced l>y C. Conseciuently the faint yellow awn appears t<^) l>e devoid of chromogen.

The oxidation and subsequent changes of the chromogenic suljstauce leading to the formation of brown pigment may be due eutirel}- to jxjst moi-tem chjinges and may have no relation to the action of a gene whatscever. If any gene is concerned, we may suppose the following possibilities.

Tlie gene C h;is the simidtiineous action of converting the chromogenic substance to the brown ])igmout in which the oxidation plays an important role.

Ol- ^\■e may suppose that another gene 0, a chroraophelein, which con- verts the chromogenic substance to the brown pigment, and the genes C and 0 are so linked each other that they may l)e considered as a single gene complex. In the awn of Oryza, and in the seed coat of Glycine as we shall see later, the chi'omogeuic substance and certain brown pigments which are the oxidation product of the former, appear to be due to the action of a single gene. Wherever the chromogenic substance is produced, it is invariably con- verted to the pigment of phlobaphene natiu-e, unless the inhibitory gene enters into the system. In this connection, it is of interest to refer the findings of Wulff, Wolff and RorcHERMAXN' and Mrs. Wheld.\le Onslow (1919),'^ who have shown in a number of cases that the reaction of direct oxidase is invariably associated with the presence of the chromogenic substance.

If the Litter assumption is adopted, the genes concerning the formation of brown awn may be designated by CO and the £iiut yellow awn In- Co.

1. WOLLF, .1., Loc. fit.

■\VoLFF, J, and Eouchekmaxx, X. Loc. cit.

2. Wheldale Onslow, JI., Oxidiziiig Enzymes. I. The nature of the " peroxidase " natur;»Uy associated with certain direct oxidizing systems in plants. Bioch. .Jour. 13 : 1, 1919.

60 I. NAGAI :

Furthsr we admit that the geue R is present üi the faiut yellow awn plant, and by the completion of the system provided by the genes CB (COR), the chromogenic substance is converted to red authocyanin, but by cR (coR) and cor {cor) the system is incomplete. The parental faint j-ellow plant may there- fore be designated by cR (coR) and the brown plant by Cr ( COr). The retl awn F^ plant is Cc Rr (COcoRr) and by selfing, the following zygotic series would arise in F, by the ratio 9:3:3:1, viz., CR {COR), Cr (COr), cR (coR ), and cor (cor ), and in which the last four would be faint yellows giving rise io 9 reds, 8 brawns and 4 fain^ yellows. The assump'iou covers the numerical ratio observed in the F^ and F^ generations.

Let us suppose in another way that the brown awn may have the genetic composition COr and the faint yellow cOR, in which 0 is the gene common to both of the parental plants. It is necessary to suppose that no red au- thocyanin should be formed liy eOR in which the chromogenic substance produced is onl}' in such a small amount that no anthocyaniu is formed fi'om it even in the presence of OR. If however, the reduction processes set work- ing by the gene complex here concerned are just as powerful as we provide in vitro by means of hych'ochloric acid and magnesium powder, even a trace of the chi-omogenic substance should be converted to the wloiu-ed anthocyaniu, for, we can readily detect even a fa-ace of the clu-omogenic substance (1 : 20,000) by a distinct ])ink colour by reduction.

In the case of the cross " Hanbun-nento " x " Genroku-mochi ", in which the purple and red awn are concerned, the basal system of anthocyaniu formation CR is complete in both of the pai-ental plants. If we let the gene R' convert the red anthocyaniu to pm-^jle, the purple awn may be designated by CRR' and the red by CRr'. The designation CRR' may be substituted by a single letter, say P, and CRr' b}- j). Since we observe the stages of the change, chromogen — » red pigment — * purple pigment in the plant, CRR seems to represent the actual phenomena occurring in the sporophyte.

In a uuuiljor of cases reported, the puiple colour is dominant o^or the red.' In AniirrJihiura ]Miss. Whedale (1914) '.found that orange authocj-auin

1. See summary in Uatk^ov, \V., Mexdel's Principles of Horcility. Third edition. l!li;i. Whbldale, JI., Anthücynn risnients of Plants IDlfi.

2. Whei.dale, .M., Oar Present Knowledge of the Chemistry of the ilendelian Factors for

Flower Colour. .lour. Genet, -i : S, IDl-i.

A tiENETICO-PHYSIOLOGICAL STITDY OX THE FOIiMATION KTC. 61

\\;is the dorivatix'C of !ii)iji;eniu jukI rose dove Jiml luiirjeuta from lutoi>liii. Tui' factors are uecessan- to convert the cliroiuogeii duteolin) to mageuta but oulj- one is essentiiil to rose dore. Thus

UyiiRRl^i 7JIIÜRHU11 White

YYIIrrU} Ivorj- (apigeuiu)

YYitriib Yellow (luteoliu)

YyilRM Orange

YijiiRrliJ) Crimson

YijIiRrhl) Kose dore

YijRRrUn Magenta

in wliicli / is a dominant ivov}' factor, inliibiting the formation of luteolin. Y is the factor for yellow whicli is due to apigeuiu, R and B are the fsictors vhicli convert the chromc^n to anthocjaniu.

The case of Liiiaria maroccana, studied by CORREXS (1912)' was as foUo^vs : Ked is dominant over white, and purple over red. At presant his inteq^retation of the above ease is hardly to Ije maintained. He dwelt upon the fact tluit authocyauin is red in acid aud blue in alkali aud tried to inter- pret the factor dillerence of purple and red by that of acidity- in the ceU sap. But such seems to be unlike the usual case. H;VAS (1916)' found that the cell sap containing blue anthocyauin is acid or neutral but rarely alkaline if acidity is determined by the hydrogen gas chain.

WlLLSTATTER attempted to explain the variation in the colora* of cyanidin from red to blue In- the quinonoid sti'uctxu-e of the molecule. The phelolic character of the benzene i-ing allows the formation of salt with alkali, the red is the acid salt (oxonium salt) the blue is the jOTtassium or metallic salt (alkali pheuoLite) aud the violet is the anhydride of the pigment. But his hypothesis hardly explanis how the deep lilue anthocyauin can exist in the cell sap of the pLant which is acid in reaction and further that an addition of alkali destroys the blue anthocyauin rather than deepening the colour.

A new explanation of the causes of variation in the colour of flower's has been put forward by K. Shibata, Y. Shib-^ta, aud I. K-AShiwagi (1919)' based

1. CoiuiEKs, C, Die Neue Vererbungsgesetze. 1912.

2. Haas, A. E., The AciiUty of Plant Cells .is shown by X.ntur.nl InJic.itors. Jour. Biol. Chcm. 27 : 233, 1916.

3. Shibata, K., Shtbata, Y., .-»nil Kashiwaci, I.. Loc. cit.

62

I. NAGAI :

upon esparimeutal evident'?, tliiit the metal organic or complex compauncls of rednced flavonol gliicoside

/

(sugai-J

3IeX

O

/\/\_./ v

I

V OH

-OH

J

MeX.

are the most important factor in the production of flower colom-s. The blue anthocyanins ai'e complex compounds of reduced flavonol glucosides, which possess several hydi-oxyl gi-oups belonging to the flavonol nucleus besides those of sugar molecules, and the metal with which they are co-ordinated is probably calcium or magnesium, for salts of these metals are always present in the plant cells. The violet, violescent red or red pigments ai-e either the analogous metallic complex compounds of flavonol glucosides, wliich contain fewer of the auxoclirome hydi'oxyl gi'oups or are a mixture of the blue pigments and their decomposition products bj- excess of acids, i.e., the red oxonium salts of 11.

WiLLSTATTEE.

It is likely to be inferred then that the purple and rod anthocyanins formed in the awn and in other parts of Ori/za are the derivatives of the sarte chi'omogenic flavonol glucoside and the pm-ple is the complex salt of the red anthoc3*anin which is formed at first in the cell b}' the reduction of the clu'omogenic substanc3. The latter part of the changes may be due to the action of gene B. Indeed, it can be obseiTed that the extract of the shglitly red colom-ed awn yielded a more intense red colour by reduction than that produced by a simple addition of hydi'ochloric acid, and the faintly coloured exti'act of red and purple anthocyanins that is due to isomerization, attained a bluish hue by tlie addition of ZuCL and a reddish hue by CaCL. The gene R' may therefore be i-efen-ed to the agencies which set free the reactions leading to the formation of a pm-ple complex salt fi-om red anthocyauin ■with tlie existing metallic salts in the cell.

With respect to the colour of the paleas, tlie following hypothesis may be pravide«.!. Let J] be the gene for the brown gixjuud-colom-, l> for tlij non

A GENETICO-PHYSIOLOGICAL STUDY UN THE FORMATION ETC.

63

In-owu. The c-hromogeuic substiiuce of the bruwu pigineut in tlio paloas was tested at the time prcvions to pigiueutatiou but failed. Therefore it is evident tliat the chromogenic substiuice of the browu pigment iu the palea is different frnm that of the awn wliich gives risa to tlie red anthocyauin by tin action of the B gene. But B has a specific relation to the formation of pm-ple pigment iu the epidermis, inasmucli as the full eolnnr in the purple palea is always associatefl with the brown gi'ouud-colour, and tlie loc^zed purple excludes the latter. It appears therefore that either />' h-is a simultanwni^ acti<jn on extending the purple pigment to the euUre surface of the palea or that the phenomenon is due to another gene which is completely linked with JJ.

The formation of any pui-ple pigment in the palea appears to be due to another gi"Oup of genes which are simihir in kind to that which afiected the awn, inasmuch as the pm'ple awn is completely linked with any purple present in the palea but no red ever occurs in the latter. Let P' be the gene for the presence of pui-ple pigment in the palea, and p' for the absence of the same. Further admit that P' and P are Hnked. Thus

" Haubuu-nento " ppp'p'BB Awn red, paleas brown.

" Genroku-raochi " PPP'Plh Awn pui-ple, paleas purple localized.

F, " Hanb." X " Genrok." PPPp'Bh Awn piu-iJe, paleas pm-ple.

The Fl plant is heterozygous to P' and B, since P and P are linked. In F. we es].iect the following genotypes to occur ;

Genotype	Plienotype Awn Paleas	Designation
PP'B	Purple	Self purple viixh. brown	PP
PP'b	••	Localized purple withoiit brown.	P
PP'B	Red	Brown	B
pp'b	"	Yellow	Y

We should expect therefore the following scgi'cgatiou in F^ -with respect to the colora' of paleas.

Genotype of K plant

PPBB 1

Segregation in F^ Kntio.

PP Const.

64			I.	XAGAI :
Genotype of F.^	l)lant				Segregation in t^^	Katio.
P'PBh	2				FP:P	3:1
Fp'BB	2				PP:B	3:1
Fjj'Bh	4				PP:P:B: Y	9:8: 3
FFhb	1				P	Const.
Fp'hh	2				P: Y	3:1
p'p'BB	1				B	Coust.
p'li'Bb	2				B: Y	3:1
p'p'hh	1				Y	Const.
Tlie actual result olitaiued is	giv	eu	iu the folL:>wiDg table.
			Table 28

Showing the result obtainecT in F^ of the cross

" Hanbiin-nento " x " Genrolai-mochi ".

See also Table 33.

Segregation	Eatio	No. of families observeil	Exiiected	DifE. (ob.— ex.)
FP	Const.	1	4.187Ö	-3.1875
rP:P:B:Y	9:3:3:1	15	16.7.500	-1.7500
PP:B	3:1	9	8.3750	+0.62.50
PP-.P	3:1	10	8 3750	+ 1.6250
P	Const.	7	4.1875	4 2.8125
P:Y	3:1	13	8.3750	+ 4.6250
n	Const.	3	4.1875	-1.1875
B:Y	3:1	5	8.3750	-3.3750
y	Const.	4	4.1875	-0.1875

If we legai-d the gene P=CRB', x>=CRr as ah'eadv discussed, P ma^• also l)e regarded as a complex of genes CEE' l>ut })' must differ fi-om p, inasmuch as in the paleas, piu-ple and non piu^Je are allelomorphic, but no red ever occurs, whereas in the awn, pnrple and red are allelomorphic to each other. It can be imagined that in ^j', err' behtive as a single unit, allelomorphic to the complex CRR'. Else F" and p>' maj- dilier oulj in C bnt contain E and R' providing that cRR' forms no authocjauin.

A OENETICO-PHYSIOLOGICAL STIDY ()X THE FoRJ[ATION ETC. f)5

With respect to the coloiu- of the gi-jiiu, let two piiirs of geues 0' o' iiud C c' be responsible. The chroiuogeu C pixxluces the cluxjinogeiiic substauce P in the testa and in pericarp, and c' produces practically uouo. The gene 0' converts the cliromogeiiic substance to a i-eddish bix)wu pigiueut especialh- in the testa and o' to a less extent. Thus bj- CO' the reddish brown gi-aiu is formed and by Go' the yellowish brown gi-ain. Tlio ]3arontal reddish brown- gi-ain plant may he assumed to Ixi CO' aud the " white "-gi-aiu plant c'o'. Tlie 9:3:4 ratio would arise in F, by selfiug the F^ plant Cc'O'o'. Thus

GO' 9 Keddish brown

Go' 3 Yellowish bro«n

c'O' 3 "White"

c'o' 1

With respect to the colour of the jialeas, the segi'egation in F^ -was similar to that of the cross " Hanbun-nento " x " Geuroku-moclii ", so we maj- assume that the analogous genes are concerned in the cix^ss " Otsubu " x " Haguro " in which the gi-ain colom-s ai-e studied. But one of the genes for the gi-ain coloura pix)bablj- 0' is completely linked with the gene for the purple colour of the palea. Suppose 0' is linked with P but C is independent of the latter, we expect the following segi'egatiou in i^. .

BPO'C 27 Awn pm-ple, palea self purple, gi-ain reddish brown.

„ „ " white "

„ localized pm-ple reddish brown.

„ „ " wliite "

brown brown yellow, brown.

„ „ " wliite "

„ yellow yellow, brown.

" white "

If reddish bivmiis, yellowish bl•ü^\■ns and " whites " ai'e added together iiTC- spective of the colour of the paleas, we obtain 3G : 12 : 16 or 9:3:4 ratio. The ratio of coloured to non coloui*ed grains in each colour tyjies of the paleas is 3 : 1 showing tliat the gene C is indei^endeut of P' and 0'.

As we have already seen, green and yellow seed coat of the soy l)eans are dominant over bhick, brown and buil'. The gi-eeu and yellow contain

BPO'c'	9
bPO'G	9
bPO'c'	3
Bp'o'C	9
Bp'o'c'	3
hp'o'C	3
hp'o'c'	1

66

I. XAGAI ;

practically uo oln-omogeuic substance whereas the rest of them is prominent. This is t) inrlicate that a dominant inhibitor is present in the green and yellow. Let G be the gene for the gi-een colom' and g for the yellow. Fur- ther we assume that C and c are the chromogeus. Tlie amount of chromo- genic substanci produced by c is less than that produced by C. Let 0 be the clu'omopheleiu which converts the chromogenic substance to brown pigment and o to reddish brown. It is assumed that the same chromogenic substance is converted to the purple anthoeyanin by B but not by r. The gene I inhibits the fuU development of the pigment in the seed coat and by i no such effect is done. The iuliibitory action of the gene / seems to extend to the action of C and R. Accordingly the different colour tj-pes may be de- signated as follows :

COGRI	Bine tinged gi'een
COfjRI	yellow
COGrI	Non tinged green
COgrI	yellow
COGRI	Black (gi-een hypostatic)
COgRi	,, (yellow hypostatic)
COGri	Brown (with gi'een tinge)
COgri	„ (without gi-een tinge
Cogri	Eedchsh brown
cOgri	Buii"

The genetic composition of tlie parental plants of the cross between "Haiü-o" and "Tanishi" would be CCOOGGmi and CCOOggRRII re- spectively. Tlie F^ plant is therefore heterozygous to three genes Gg Rr and /(■ but homozygous to C and 0. The different phenotypes appeared in F. are due to the recombination of these genes. Thus :

	Blue tingetl	Black	Non tinged T ■ ■	Brown	Total
Yellow Green	COgEI 9 COGBI 27 Anthocynnin p	COgRi 3 COGRi 9 resent 48	COgrI 3 COGrI 9 Antliooyanin	COgri 1 COQri 3 absent 16	16 48 6i

A GENETICO-PHYSIOLOGICAL STUDY ON THE FORSrATION ETC.

G7

la bliu^ks, yellow, greon or brown is completely covei-ecl, but iu brown gi-een colom* in visible, hence blue tinged gi-eens, blue tinged jellows, non tinged gi-eens, nou tinged yellows, blncks and bi-owns arise by the ratio 27 : 9 : 9 : 3 : 12:4. If we denote alwve plienotypes by B.T.G., B.T.Y., (i, Y, BJ, and Br. respectively, the following segregation is expacted in the F^ generation.

	F, plant.		Segi-egation in F,	Ratio.
B.T.G.	G GBR II	1	B.T.G.	Const.
)j	GGlUii;	2	B.T.G. -.ni.	3:1
))	GGRrIf	2	B.T.G.: G.	3:1
»J	GGRrll	4	B.T.G.: G: in -.Br.	9:3:3:1
))	GijRRII	2	B.T.G. :B.T.Y.	3:1
»J	GgRRIi	4	li.T.G.: n.T.Y.: Bh	9:3:4
»	Ggllrll	4	B.T.G.: B.T.Y.: Y	9:3:3:1
)>	GgRrll	8	B.T.G. : G. : B.T.Y. : Y : Bl : I	,V. 27:9:9:3:12	:4
Bl	GGRRÜ	1	Bl.	Const.
39	GGRrll	2	Bl : Br.	3:1
9)	GtjRRll	2	Bl.	Const.
99	Gf/Rrli	4	Bl. : Br.	3 : 1 (12 : 4)
JJ	ggERli	1	Bl.	Const.
J)	ggRrli	2	Bl. : Br.	3:1
G.	GGrrll	1	G.	Const.
99	GGrrll	2	G : Br.	3:1
>>	Ggrrll	2	G : Y	3:1
>)	Ggrrii	4	G: Y:Br.	9:3:4
BT.Y.	ggRRII	1	B T Y.	Const.
99	ggRRIl	2	BT.Y.:Bl.	3:1
99	ggRrll	2	B.T.Y. : Y	3:1
99	ggRrll	4	B.T.Y: Bl: Y. : Br.	9:3:3:1
Br.	GGrrll	1	Br.	Const.
i9	Ggrrii	2	Br (gi'eenish) : Br.	3:1
l>	ggrrll	1	Br.	Const.
Y.	ggrrll	1	Y.	Const.
»»	ggrrii	2	Y : Br.	3:1

The observed data showed a clos3 approximation to the above expectation. See the table below.

68

I. NAG.U :

Table 29.

Showing the result obtained in F.i of the cross

•■ Tanishi " X " Haiiro." St-e alsii Tnbk' ?,\

Fi phenotype	F> gonotype	"So. of families obsen-ed	expected	difference
B.T.B.	GGRRII GGRRIi GGRrll GGRrii GgRRII GgREIi GgRrll GgBrIi	2 1 3 3 3 1 4	0.594 1.187 1.187 2.375 1.187 2.375 2.375 4.751
		17	16.031	+0.969
SI.			1 f.	2.375 4.751
		7	7.126	-0.126
G.	GGrrll GGrrIi Ggrrll Ggrrii	2 1 3	0.594 1.187 1.187 2.375
		G	5.343	+ 0.657
Ji.T.Y.	ggRRII ggRHIi ggRrll ggRrli	1 1 1 3	0.594 1.187 1.187 2.375
		6	5.343	+0.657
Y.	ggrrll ggrrli	—	0.594 1.187
		—	1.1781	-1.178

A GEN*ETTCO-PHYSIOLOGICAL STUDY ON THE FORMATION ETC.

69

Fj phinotyiK;	Vi gtnotyiH!	DO. of famiUesi olMerv«!	expected	difference
Br.	Ill	1 1	0.594 i.lH7 11.594
		..	■J 37 ~	-0.375

In the case of the oi-oss " Waremame " x " Acliuiuuri " iu ^\ hicli biili" and black ivve ci'ossed, two paii-s of genes are concerned in the formation of four typas of seed coat, i.e., self black, imperfect black, brown and biiß". If we let a pair of genes C and c stand for 'the chromogens, imperfect black wivs fonnd to can-y no C, in spite of the fact that the gi'ound-coloiu" appeared to be brawn, as already mentioned, henca we may assume that the full development of the deep purple authocyanin pigment is only possible by the presence of C and R. The gene for the formation of the chromogenic substance may therefore have a simxdtaueous action on the formation of brown and self black from the same chromogenic substance. Here the gene G is not concerned. The parental plants may have the follovtdng genotypic com- position with resp3ct to the coloiu- of the seed coat :

Self black CCOOREiigg

Bufl* ccOOrriiyrj

The F^ plant is heterozygous to C and R, and in F^ we shoidd expect the following families :

Self black	CCRR	1	Black			const.
J>	CCRr	2	Black : 1 irown			3:1
?»	aRH	2	Black : imperf.	black		3:1
Self bhic-k	CcRr	4	Black : imperf.	black : brown	buff	9:3:3
aperfect black	ccRR	1	Imperf. black			Const.
»	ccRr	2	Imperf. black :	buff		3:1
Brown	CCrr	1	Brown			Const.
»	Ccrr	2	Brown : biiii'			3:1
Buff	ccrr	1	Buff			Const.

The actual numbers observed are as follows.

70

I. NAGAI

T.U5LE 30.

Showing the result obtained in F:\ of the cross '■ Waremame" x " Aehntauri' See also Tal.le 33.

F-2 pknt phenotype	genotype	no. oljserved	of families expect etl	diff. (ob. — exp.)
Black	CCJIF.	3	2.875	+ 0.125
„	CcUr,	8	5.7.50	+2.250-
»'	CClir	6	5.750	+0.250
„	CcUr	1.Ö	11.500	+ 3.500
Black imp.	ccRR	2	2.875	-0.875
„	ccRr	6	5.750	+ 0-2.50
BrowTi	CCrr	3	2.875	+0.125
,.	COi-r	3	5.750	-2.750
Bu£E	ca-r	0	2.875	-2.875

The difierent coloiu- types of the suv bean seed coat so fai- concerued constitute the following series when they are aiTauged according to domiuaucy : blue tinged green > blue tinged yellow > green > yellow > black > imperfect black > browns (lighter brown >deep3r brown) > buff.

It is of interest to compare tha case c)f tha say bean to that of the Adzuki lieau -nhich have been investigated I)}- Takahashi and FUKUYAMA (1917)'. They have shown that the different colour typas behaved strictly in accordance with MeudeUan principle as iu the case with thä say bean. The different types can be aiTanged, according to dominancy, as follows : Blue black > black > black flecked (" Yogore "j>bkck flecked red>greeuish grey> deep buff (" Cha ") >red (self) > red-eyed white> white.

The test for the chromogenic substances akeady romarkeil (see Table 8) shows that the chi'omogenic substance P and F are present iu the green, mu-ipe beans of all the coloured types but absent iu white which is a " warm bufl'" according to the nomenclature by RrDGWAY. White is the most recessive character in the series. In the case of the soy boau, the typas which show verp lifile chromogen content are green and yellow ■which are dominant over the types rich in the chromogenic substauca. The diflerence iu the genetical

1. 'I'akahasbi, Y. and Fvkvsatha. J. Morphological aul Genetio Stmlies on the Ailzuki-bean. Hokkai<lo Agric. Exp. Station. Japan. EeiKJrt 7. jip. Ißl (in .Tajianese).

A OENEl'ICO-PHYSIOTOOICAL STUDY ON THE FORMATION ETC. 71

Ijehavior of the cliroiuf^u coutniniug typjs iu the two species oi plauts is due to the preseuce jiud iibseiica of au iuhibitor. The authni-s showed that iu the cros3 between bull" aud white, the i^ seed was buii' aud segregated iu F. hnih, reds aud whites by a rate 9:3:4. "WTiile the ctoss batweeu deep buff and white gave deep bull* iu I<^ {F. seed) aud deep buife, I)uflfe, reds and whites in the F, generation by the ratio 27 : 9 : 12 : Id. Tlius we see that the deep buff differs fi'om buff by a single fiictor-difiereuce aud ImfV :iud i-ed also by another fjiyctor pair. We see therefore tLat the inliabitor is also present in that case but diffei-s fi-om thixt of the soy bean in such :i way that the iuhibitor in the Adzuki-bean inhibits the formation of i-edtiish brown pigment from the chromogenic substance but the action does not seem b.> extend over the formation of the chromogenic substanci^, while iu the case of the soy beau, the inhibitor iuhibits the formation of the clii-omogenic substance, so that no chromogenic reaction can be observed iu the green and yellow. In the bujßfe, the chromogenic substauce can readily be directed and if another gene, a chi'omophelein is added to it, the deep buff is produced.

The writer was able t<j test the chromogeuic substance by the material which was kindly furnished to him through the courtesy of Mr. Fckutama. Fully ripened deep buii" and buff were found to be rich iu the chi'omogenic substance P. The gi-een unripe beans born on the plants raised from the same material in the next year also gave the similar result.

Iu order to compare the genotypic compositious of the types of Adzuki and those of the soy beans, it is conveuieut to change the designations used by the authors to those proposed iu the present j'aper. Thej- gave EH/ for l)uff, Rfif f T rod, and rlif for white in which B is the gene for red, // au inhibitor and F the gene for buff. We assume thai the genes for the red" pigment are C and o aud by th3 action of an inhibitor / results in the formation of buff. The gene / only inhibits the action of o but that of (J is loft fi-ee. We have some data for believing that the reddish brown . pigment of the Adzuki-bean is the oxdidation product of the chi'omogenic substance. Tlio j)igmont is iusolublo iu stroug ucids, Inxt readily soluble iu water and especially iu weak alkalies Aielding a deep wine red colour which becomes yellow by sicid. The alkaline pigment is insoluble in ether and acatic eth3r but iu a weak acid solutiou, sparingly soluble in ether, aud by evaporating

72 I. NAGAI:

the solvent yields an amoi-phons recltlisb brown pigment. Tlie pigment may be precipitated from aqueons solution by le<id acetate.

Consequently the gene / seems to inhibit the action of oxidizing agency acting on the chi-omogenic substance. The alx)ve mentioned relation which exists in buff and red seems to be analogous to that of the case of the dominant white in Prininla sinensis. According to Keeble and Armstrong (1912y, Keeble, Armstrong aud Jones (1913)- and Keeble and Miss Pellow (1910)^ certain dominant wliites contain chromogen, which occurs in the recessive white in au extremely slight amount and the inhibitory substance which obscures the peroxidase reaction is present in the former. The buff coloured seed coat of the Adzuki-beau can be considered somewhat analogous to the dominant white in the flower of Primula and the white to the recessive white. The colour of buii" aud white in the seed coat differ shghtly from each other. The peroxidase reaction in the seed coat was also examined aud an indication to the similar relation that was observed in the flower of Primula sinensis was obtained. In the epidermis of the seed coat in which the pigment is confined, the peroxidase reaction was extremely slight in the unripe green seed of buff aud de?p bulf whereas in the white, very distinct The observation was repeatedly made Avith the material taken at the different stage of maturity. The section was placed under the cover glass with the alcoholic solution of benzidine or alpha naphtol witli a dilute solution of hydrogen fieroxide. In this manner, the direct oxidase has failed to be detected in all cases.

Tlie reddish bi-own in the soy beau (such as " Aka-nedzumime ") con-e- spouds with red in the Adzuki-bean. The gene 0 which modifies the redfhsh •brown to bro^^•u iu the soy beau corresponds with the geue / in the other, inasmuch as they suppress the formation of reddish-brown, oxidation product of the chromogenic substance, though they differ in the manner toward the formation of the chromogenic substance as already mentioned. The genetic composition of the different self coloured typos in botli species can ba expressed by the same designations iu the following manner :

1, Keeble, F. nnl Akmsteong, E. F., Loc. cit.

a. Keeble, F., Aemsteonq, E. F., an;l .Ioves. W. N., The Formation of the .Vnthooyan Figments of Flnnts. 6. Proc. Roy. Soc. London, R 87 : 113, 1813.

3. Keehle F. ami Fellow, C. White Flowered Varieties of Primula slnensus. Jour. Qeue- tios. 1 : 1, lUlO.

A GKNETICO-PHYSIOLOGICAL STUDY ON THE NOBMATICN ETC. 73

		Soy bcfvn		Adzuki-beiin
Buff		cOi-iij	AVliito	coiiij {rfh)
Ketltlisli	brawn	Corbj	Keel	Coruj (R/h)
Brown		COrig COriG	Buff Deep buff	CorLj {B/II) GOrhj {HFII)
Black		C'OKi[i etc.	];iac-k	CorLj (HF/iMC)
YeUow		COfLj etc.
Green		COrIG etc.	( ireen	COrIG etc.

Also, in the seecl coat of Pkisedm vulgaris, Phaseolus ranUißorvs^ Pisum sativum-, Viyna nngtiimlia and Vigtm sinensis^, the C(.>loured types which nre duo to anthocyauin pigments are doiaiuaut over those which ai'o duo fcj phlobapheue pigments and the latter are dominant over white types. In Pisum, the well known work of Mendel lias already shown that the brown coloured pja is a simple dominant over the colourless one. Lock enumerated the genes concerning the colom- of the testa as follows ; (1) greyish or brownish pigmentation as opposed to the absent (white) (C) (c), (2) purple spotted of bright pm-ple spots as opposed to very faint or absence of the character (S) (s), (3) the presence of maple character. Mapliug or mottling of a rich brown coloiu' as opposed to the absence of the character (31) (m).

1. Lock. K. H., SUidies in Plant Breeding in tlie Tropics. Ann. Koy. Bot. Giinl. Peradcniya. 3 : 95, 190G, Shcll, G. H., Some Latent Characters in White Bean. Science. N. S. 25 : 828, 1907. — Shtll. X yevr Mendelian Eatio and Several Types o£ Latency. .\mer. Xat. 42 : 433, 1908. — TscHEEMAK, V. E., Weitere Beitrüge ueber Verschieilenwertigkeit der ilerkmarle bei Kreuzung von Erbsen n. Bohnen. Zeit f. d. 1. Versuch. Osterreich, 1901, 641.— Tschekmak, Weitere Kreiitzim.2;s- stndien an Erbsen, Levkojen und Bohnen. Ibid. 1904, 533. — ^Tschermak, Bastardiemngsversuche an Levokojen, Erbsen, \md Bohnen mit Rücksicht auf der Faktorenlehre. Zeit.f. induk. u. Vererb. 7 : 81. 1912. Emebsos, 1'. X.. Inheritane« of Color in the Seeds of the Common Bean. ".Phaseolus \-xilgaris. Ann. Beix)rt iSebraska Exp. Station 22 ; G7, 1909.— Lcs-dbeeo, J. and Akermav, A , The Colour of the Seed in the Descend.ints of \ Natural Hybrid of Two Varieties o£ Phaseolus vulgaries. Sveriges Utsadesforeinge Tidskrift. 27 : 115, 1917. (cited in Intermit. Eeview of Sc. and Pract. of Agric. 8 : Entry 1013, 1917.)

2. Lock, K. H., The Present Stage of Knjwledge of Hereility in Pisum. .4nn. Eoy. Bot. Peradeniya. 4 : 93, 1908. Mesdel. (.i.. Versuche imler Pflanzenhybriden. Verband. Naturforsch. Verein in Briinn. 10 : 1865. — White, O. F., Researches on the 35 Factors Determining the various Characters of the Genus Pisum, Jour. Agric. Research. 11 : 166, 1917.

3. Spillman, W. J., Inheritance of the •' Eye " in Vigna. Amer. Nat 45 : 513. 1911.— SpxiJ/- MAN-, Color Correlation in Cowi)ejis. Science N. S. 38 : 302, 1913. — Harlaxd, S. C, Inheritance of certain characters in the Cowpaa (Vigna Sinensis)- Jour. Genet. 8 . 101, 1919.

74 I. NAGAI :

In ceitain varieties of Phaseolvs vulgaris, Tscherjiak ( 1912) showed that the coloured seed coat was dominaut over white, and among coloui'ed types the relation, black > violet > brown was estabhshed. Shull (1508) also found in the same x^lant, purple, brown, yellowish-brown and yellow were dominant over white. He proposed the following genetic composition for the diflerent self coloured types :

Brown and yellow Pbm

Black PBm

White 2^B3I

in which P is a gene for the pigment, B the modifier of the pigment, and M the mottling gene. P may correspond with CO or C and B witli 7i in our case.

In dealing with the colour of patterns of the seed coat of figiuz ungnl- culata and Vigna sinensis Spillman (1913) and Haeland (1919) respectively foimd that the solid coloured types were dominaut over the mottled and less coloured ones. The latter author showed als > that Uack was doroinjxnt over brown, and brown over red. The brown was completely dominant over red in Fl and brown, maroon, and red arose in the F-, by the ratio 12 : 3 : 1. These colours are of the phlobaphene nature and no anthocyanin is concerned except black. The genetical behaviour of the phlobaphene colour types in this plant is quite similar to tlia": wliich we have seen in Adzuki and soy Ijeaus. In all these casas the more intense reddish-brown is recessive to less intensely colovu"ed types.

In Zea 3Iays, East and Hayes (1911)' showed that the dark-red pericai-p was a simple dominant over -white. The coloiu- of the pericarp is due to the pigment belonging to the phlobaphene group. The pm-ple and red aleurone coloura are due to anthocyanins. The formation of the anthocyanin pigment in the aleurone cells is in certain cases, governed by the genes (C. B, P of East and Hayes, C, R, A, Pr. of Emerson) which are apparently similar iu kind as those met iu the case of Lalhijrus, AntirrJdnviu and Onjza.

1. East. I'. 31. nnl Haye!, H. K., Inheritance in Mixize, Bull. Conn, .\grio. Exp. Station. 1C7, 1911. — Emebson, E. a., A Fifth Tair of Factors, .1 '(. for Alerirone Color in Maize, and Its Kolation to the C c and R r Pairs. Cornell University Agrie. Exi). Station Memoir IC, ISilS.

A GENETXCO-PHYSIOLOGICAL STUDY ON THE FORMATION ETC. 75

Thus we see that the pigment yielding mechjiuisiu in the seed coat of different species of jjlant fcilLs in general under a similar category, imrticnlarly in the sewl coat of tlie legumes.

4. Discussion.

To a gene we imply a specific protopLismic eutiti,- ^\•hich sets up tlio biochemical apparatus in the sporophytic cells and to the end product of the reaction psrformed b}- the mechanism so set up, we refer a chaiucter, morpho- logical and physiological. Therefore, even we infer a gene to a character, that gene itself may liavc no direct relation to the character. A catalyst does not appear in the final prfxluct of a chemical reaction, but may alter the ralocitj- of the i-eaction and sometimes change the position of e^uihbrium to bj attained.

When such agencies or genes are f)aired forming an aUelomorj)h, and they sogi-egate in a normal way, we can deduce the relation between the character and the gene by the numerical ratio of character that is required by the supposed genetic entities. We disregard the Inochemical processes involved in the changes which are set up by the gene to bring about the equihbrim, of which state we perceive the character. Dynamicsüly viewed, however, the possibihty is not excluded even in such a case in which a single allelomoi-phic character-diflfereuce is due to more than a factor-difference. Supposing the cliange A-*0 in which the substance A undergoes certain ch.inges to form the substance 0 which may be regarde 1 as a single character in the Mendelian sense, such as a purpb pigment in a certain organ in the plant. A-vO reaction would appear to be a single change when the initial and the end product alone is considered, l)ut it may involve the catenary changes A~>B-^C^B-^0. Such complex changes are Ukely to occur in most of the biochemical processes like respiration and photosynthesis which seem comparatively simple when the initial substance and the final product alone are considered.

If we consider an imaginary in3l:ance in which A -^B, B -» C, C-*D, and D - 0 reactions are involved in a whole change A^O, and these separate changes are governed by the respective genetic entities, yet they ai'e not

76 I. XAGAi :

separable at the time of synapsis. The character which is due to the end product of the final change, therefore woidd appaar as due also to a single genetic entity. Suppasiug that the purple authoeyanin is produced by C-*B-*P changes and three separate genes are actually taking pai't to biing about complex chemical changes. But if these genes are linked, or so to say, form a complex, and do not separate in gametogenesis, they may weU be considered as a single entity and can be substituted by a single designation to express the genetic entities to a given character. When they separate fi-om the complex by any cause, a supposed single imit character M"ould appear to be constituted by more than a single gene.

The separation of genes from tha complex may take place either by hybridization or by unknown internal causes, and of the latter cases, we call mutation.

It is a comparatively simple matter to determine the number of genes concerned with given characters by h^'bridization experiments when the contrasting characters are distinct and the segregation in the oflfepring of the hybrid is sharply defined. But it is exti-emely difficult to interpret those genes in terms of biochemisti-y or physiology. We are likely to fall into the danger of providing a superficial analogy and di'awing sweeping CDnclusion by confusion of the geuetical data to those of physiology.

lu the case of the formation of anthocyanin and phlobaphene pigments in the plants studied, the genes C, 0, R and P appear to govern certain gi'oups of biochemical reactions in the sporophytic cells more or less in a distinct manner, yet we must have great reserve in referring these genes to any physiological factors. It is time that the paroxidase coexists with the pigment, and the normal oxygen relation is essential to the formation of the pigment, but these facts prove in no way to jillow us in intei-preting the complementary gene of the colour producing system in plant is exclusively relating to peroxidase. Even in the case of the formation of brown and reddish brown pigments, in which the oxidation of the chromogenic substance is an essential change, the direct inference of the geue to peroxidase or oxidase may deserve serious consideration.

The formation of brown plant-pigments (plili)baphenes) resembles, as we have ah-eady seen in the precethug pages, that of melanin pigments in animals

A GENE'nCO-PHISIOl-OOICAL STUDY ON THE FORMATION ETC. 77

ill some resjÄcts. Certain antliüi's go so fiir as to regard the browuLsIi pigments in the seed coat of the legumes as a soii; of pbiut melaniuic pigment.'

■\VRiGirT (1917)', proix)sed an hypothesis regarding the colour inheritance in Mammals. He pro^xises fii3t, that melanin is produced l\v the oxidation of certiiiu products of protein metalxilism b}- the action of sfÄcific enzymes; second, that this i-eaction takes place in the cytophism of cells probably liy enz3-mes secreted by the nucleus ; third, that various chromogens are used, the jiarticular ones oxidized depandiug on the charactere of the enzymes present, and finally that hei^editaiy difference in colour are due to hereditary difierences in the enzyme element of the reaction. It is supposed that color depends on the rates of production or of potency of two enzymes. Enzyme I is essential to the production of any colom-, but by itself only produces yellow. Euz^-me n is supplementary to enzyme I, jiroducing no effect In- itself. The compound euzj-me I — 11 is also more efficient than enzyme I in another way. It pr(.)- duces sepia pigment even when enzyme I is at too lo\\- a potency to produce anj- yellow by itself. Above th^ level at which enzyme I produces eflects, the enzyme I and I — II, complete the oxidation of chromogeu.

Regarding the place of the enzyme reaction to the chromogenic substance, his liyix)thesis may be refen-ed to the -view of Unna (1913)' in which he maintained that in the tissue of the animal skin, the plasma is the reduction phxce (" Reduktiousort ") and the nucleus, the oxidation place (" Sauei-stoflbrt "). In plants, however, Schneider (1914)* coiUd not establish Unna's ■\-iew.'''

If the mitochoncWa is tha seat of the pigment synthesis as Guilliekmond

1. JIaxk, a.. Coloration of the Seed Coat of Cowj^eas. Jour. Agrip. Eesearcb. 2 : 33, 1914. The substances known as " Phytomelan " are, however, different from phlobaphenes. See Dafekt, F. AV. and Miklauz, R., Untersuchungen ueber die kohleähnliche Masse der Kompositen. I. Denlcschr. d. Kais. Akad. d. Wien. Bd. S7, 1911. Cited in JIolisch, H.. Mikrochemie der Pflanze. 1>. 319, 1913.

2. Wkight, S., Color Inheritance in Mammals. Jour, of Heredity. 8 : 224. 1917.

3. CssA. P. G., I3icxhemie der Haut. Jena. 1913.

4. ScHXEiDETi, H, I'eber die Unnaschen Methoden zur Feststellung von Sauerstoff-und Keduktion-Orten u. ihre Anwendung auf Pflanzliche Objekte-Benzidin als Reagens auf Verholzung Zeitsch. f. wi.s.s. Mikro. Tech. 31 : .51, 1914. a. — Sohsetdeb. Neue Studien zur Parstellimg der r.cduktions u. Sauerstofforte der Pflanzenzelle. Ibid. 47S. 1914. b.

5. Cf. OsTBKHAUT. W. J. V.. The Role of the Nucleus in Oxidation. Science. N. S. 46:3G7, 1917.

78 I. NAGAI:

and othei-s' have reporbed, the reaction place saems to be chiefly located in the cytoplasm. All the genes must be retained in the nuclear substance of the spjrophytic cell in some latent state, and the reaction done by them iu the cell to produce the pigment must be realized by some sort of substances deiived fi'om the nucleus. The actual relation between the substance of genes in the nucleus and the mechanism iu the cytoplasm conditioned by the former, is known to none of us. It seems therefore altogether premature to speculate, as certain biologists might propose, that the gene itself is the enzyme. Even in the pure chemical field, we do not know as yet the exact chemical nature and the mode of action of enzymes.

Ill Summary and Conclusion.

In a number of species of plants examined, two groups of pigments anthocyanins and the reddish brown pigments (phlobaphenes) can be traced to the chromogenic substances, previous to their formation. In certain cases, both of the pigments can be formed from the same chromogenic substance by the action of various complementary pigment-yielding agencies. ^

The chi'omogenic substances can be identified to two gi-oups of aUied sub- stances, one of which is designated as the chromogenic substance F which includes the glucoside of certain flavones and flavonols, and the other, the chi'omogenic substance P of which the chemical nature is yet unknown.

Evidence is given to show that certain brown and reddish bro-\vTi pigments (phlobaphenes) are the oxidation products of the chi-omogenic substance P and F.

Certain anthocyanins are completely decolorized by the action of oxidizing enzymes.

Certain flavones, flavonols and their glucosides yield a chai"acteiisti(! oxida- tion colom- by the action of oxidizing enjymes.

1. Gtjtlliekmont, .\., SvTr la formation de r.mthocyime an siin ilcs mitochondries. Comp. Kcnil. Acad. Sei. Paris. 156 : 1921. 1913. — GriijUEBMONr), ^oiivelles rechercbes oyfologiqnes siir In formation des iiigments anthocyaniques. Ibid, 157:11)00, 1913.— GriLiiERMOND, Quelques obser- vations cytologiques sur la mode de formation des pigments antbocyaniques dans les fleurs.

Ibid. 101 : 494, 1915 Guilijekmond, Hecberches cytologiques sur la formation des pigments

antbocyaniques. Kev. Gene. Bot. France. 25:295, 1914. — MonE.U"x, F., Loc. cit.— Mira.sde, M., Observation sur le %'iTnnt de la formation lytologique de 1' anthocyanine. Comp, llend. .\cad. Sei. Paris. 103:368, 1910.

A GENETICO-rHYSIOLOOICAL STUDY ON THE FORMATION ETC. 79

The anth'X^yauin pigmeut is the reduction pivxltiet of the chromogenic siiI)st;iiKv wlioii tlie cbiMinogeuic substance F alone is concerned, but the other jxissibihties are uut exchided when the clu'omogenic substance P is concerned.

Wlien tin' iMiuplete system is hud ilowu in tlie sporophytic cells by the combination of the sepvrate components which are retained by the specific genetic entities, anthocyanin pigment is formed in tlie awn, and glumes of Onjza saiiva. Hence In- a proper crossing, the awn nf tha h3brid plant be- tween two races which lack the pigment, forms anthocyanin.

A linkage relation was observed between the pm-ple colour in the awn and the reddish lirowu colour in the testa in the varieties of Ori/m saiiva stiKÜed.

The brown pigment of the awn of Or/j::a is due chiefly to the oxidiition pnxluct of the chromogenic substance F, and th.it of the testa is due chiefly to that of the chromogenic substance P.

The coloiu'ed stigma, pm-ple awn, palcas, and striped leaf-sheath which are due to the presence of anthocj-auiu are inherited in a gi'oup, and in the pxleas, the solid pui-ple is linked with the brown pigment wliich is formed at the uuderlj-ing tissue of thi same organ, while the localized purple repells the latter.

The brown and reddish brown pigments and the pm'xJe anthocyanin formed in the seed coat of GIi/ci?ie soja are derived chiefly fi'om the same chromogenic substance belonging to the gi-oup of the chromogenic substance P.

The formation of these pigments as well as the chromogenic substance is entirely or partiall}- suppressed b}' the action of dominant inhibitors.

Cei-taiu genetic phenomena relating to the colom-s of the seed coat are studied. The following is the list of characters studied, arranged according to dominancy in the ascending order. Blue tinged green > blue tinged yellow >. gi'een> yellow > black > brown (hghter brown>reddish br(jwn)>bufl'.

Acknowledgement.

The writer -«vishes to express his most hearty thanks to the Professors K. SiDBATA and Y. Shibata of the Imperial Univei-sity of Tokyo for the en-

80

I. NAGAI :

couragement, suggestion, and facilities given liim for caiTving out certain pliTsiological experiments. Without tbeiu, the study %\ould not have been able to reach the present state. His sincere thanks are also due to the Authority of the Imperial Agiicultm-al Espsriment Station who has given him permission to take oppoituuities of conducting this investigation. To Messers. T. NiBE, Y. Kakizaki, T. Tajiba and J. Tsukamoto and other colleagues and hiends who have heli:)efl the writer in various ways, his cordial thanks are extended.

Table 31.

Showing the result obt.iined in F^ of the cross " P.iildcoto " X " Togo."

F^ family no.	Awn of F^ xjlant	Kea	Brown	Faint yellow	Total
14	Brown	36	—	—	36
29		96	—	—	96
38		106	—	—	106
i	„	15	—	9	24
9		21	—	10	31
19	„	13	—	i	22
24	"	46	—	21	67
33	,.	49	—	11	GO
35	„	76	—	27	103
31	,.	57	—	21	78
2	„	37	20	—	57
6 •	f	19	2	—	21
10	,.	63	24	—	87
12		47	18	—	65
16	,,	65	31	—	96
17		46	20	—	66
io	,^	20	9	(1)	30
32		75	22	—	97
35	1»	16	10	—	26
34	,,	. 4S	22	—	70
3	,.	35	15	14	64
7	»»	49	15	32 *	96
8	■ .,	52	17	23	92

A GENETICO-PHTSIOLOOICAI, STUDY ON THE FORMATION ETC.

81

f J family no.	Ann of Fj plnnt	Red	Brown	Fnint yellow	Total
	11	liro^-n	45	U	ir.	75
	15	„	33	15	ir.	64
	18	„	33	111	15	58
	21)	1»	111	11	4	25
	22		2S	15	20	63
	23	.,	3n	19	13	62
	25	„	]'.)	111	5	31
	2r.		58	IS	22	98
	2S	„	41	(i	ir>	66
	1	.,	—	47	17	64
	13	»»	—	43	18	61
	21		—	64	—	64
	37	„	—	G9	—	69
	39	Faint Yellow	—	—	101	101
	40	„	—	—	63	63
	41		—	—	89	89
	42	.,	—	—	63	63
	43	?»	—	—	66	66
	44		—	—	67	G7
	45	„	—	—	33	33
	40	„	—	—	98	98
	47	„	—	—	98	98
	4S	„	—	—	101	101
	49	..	—	—	1)7	97
	50	••	—	—	31"	31

Table 32.

Showing the result obtained in Fj of the cross " Kurafusagi " x " Xiigoynsliiro ''

F3 family no.	Awn of Fj iilant	Hoi	Brown	Faint Yellow	Total
31	Brown	61	—	—	61
42	„	91	—	—	94
44	♦>	95	—	—	95
45	»	62	—	—	62

Ü'Z	I	. NAGAI:
F.J family no.	Awn of f J plant	Keel	Brown	Faint yellow	Total
80	Brown	57	—		57
82	„	100	—	—	100
84	. ..	96	—	—	96
88	„	63	—	—	63
2	..	67	29	(1)	97
3	•.	48	13	<1)	62
22	„	49	16	—	65
23	<•	67	28	(1)	96
36	„	67	30	—	97
49	.,	67	32	—	99
57	„	66	27	—	93
58	,.	51	21	—	72
60	■.	75	25	—	100
C2	.,	56	13	—	69
(U	„	45	15	—	60
69	,.	66	31	—	97
73	„	65	23	—	88
74	„	75	18	—	03
77	>»	68	29	—	97
83	„	74	17	—	01
86	„	33	13	—	46
21	„	23	8	—	31
15	„	44	(1)	15	60
32	„	46	—	16	62
63	„	19	—	11	30
85		47	—	18	65
89	„	51	—	12	63
91	„	28	—	4	32
93	„	28	—	5	33
94		30	—	4	34
97		25	—	11	36
98	,.	50	—	12	62
102		71	—	13	84
103		67	—	26	93
105	,.	72	—	25	97
1	..	27	22	11	60
5	„	32	6	24	02
10	. „	38	22	18	78

A GENETICO-rilYSIULUGICAL STUDY ON THE FOinrATION ETC.

83

l'\ family no.

13

18

19

21

25

27

'29

33

34

35

39

41

43

50

52

53

59

65

75

76

79

87 101

92 107 108 109 110 4

6

7

9 28 3) 37 38 44 70

Awn of F.^ plnnt

Brown

46

50 55 10 19 !) 23 43 I'J G2 42 43 33 00 19 40 43 29 34 34 32 4S 58 64 42 50 54 27

Brown

21

18

13 5

23 2

12 8

23

12

21

21

14

19 8

14

20

14

K)

10

15

21

19 8

11

13 18

13 44 G2 72 26 68 73 77 7G 29 67

Faint yellow

33

28 31

3 2G

5 24 16 29 22 24 26 16 17

G 17 25 22 22 15 14 26 24 19 16 32 21 25 14 22 18 10 25 2T 21 23 11 28

Total

100

96

99

18

98

16

Gl

67 101

96

87

93

G3

92

33

71

78

65

66

65

Gl

95 101

91 09 IUI 96 G5 58 84 90 36 93 93 98 99 40 95

84

I. XAGAI ;

F3 family no.	Awn of F2 pl*™t	Kea	Brown	Faint yellow	Total
71	Brown	—	78	23	101
78			41	20	Gl
81		~~	38	15	53
90			63	33	96
56		—	39	9	45
8		—	96	—	96
12		—	13	—	13
14		—	28	—	28
40		~	63	—	63
47			57	—	57
71			91	—	91
111	Faint yellow	—	—	64	64
112		—	—	33	33
113		—	—	63	63
114		—	—	99	99
115		—	—	98	98
116		—	—	101	101
117		—	—	99	99
118	"	—		98	98
119	»»	—	—	64	64
120	„	—	—	99	£9
121	.,	—	—	87	87
122		—	—	ICO	1(10
123		—	—	94	94
124		—	—	33	33
125		—		92	92
126		—		85	85
127		—	—	01	61
128			—	97	97
129	1	—	95	95
130	—	_	82	92
131	—	—	97	97
132	—		86	86
133		—		90	96
13t		—		(U	64
135		—	—	65	65
136		—	—	C2	02
137		—	—	89	89

A OENETICO-PHYSIOLOGICVL STUDY ON THE FORMATION ETC.

85

i'", family no.	Awn of Fj plant	Re<l	Brown	Faint yellow	Total
138	Faint yellow	—	—	61	CI
139	•»	—	—	33	33
140	•.	—	—	34	34
141	.•	—	—	34	34
142	»»	—	—	38	38

Table 33.

•Showing tlie result obtain«! in F^ of the cross " Hanbiin-nento " x " Genroku-mochi "

	I'aleas of J'"j plant	.\.wn paleas self P	jurple paleas loc. P	Awn	red
i'3 family no.	paleas brown	paleas f . yellow	Totals
2	Self purple	58	10	—	—	74
4		71)	21	—	—	91
18		75	21	—	—	96
•u		72	22	—	—	94
25		54	11	—	_,	G5
30		74	21	—	—	94
42		51	12	—	—	63
03		0	2	—	—	8
GO		11	4	—	—	15
C8		20	8	—	—	34
10		74	—	21	—	95
13		02	—	34	—	96
15		55	—	12	—	67
22		78	—	20	—	98
37		CI	—	21	1	82
45		65	—	21	—	8';
47		04	—	32	—	90
48		71	—	27	—	98
50		25	—	9	—	34
1«	Local, purple	—	48	—	15	03
20		—	09	—	19	88
23		—	02	~	37 (2/	101

* Not recordetl.

86		I. NAGAI :
Fy family no.	Paleas of F^ Plant	Awn purple Palea.s Paleas self r. loc. P.	Awn palea.s brown	red paleas f . yellow	Totals
27	Local. Purple	—	44	—	17	61
31	»>	—	57	—	32	89
39	»»	—	70	—	26	96
44	)»	—	77	—	27	1C4
49	„	—	77	—	23	100
53	»>	—	2.)	—	8	33
57	,,	—	73	—	5	78
58	it	—	20	—	14	31
59	„	—	59	—	14	73
64	„	—	40	—	2'i	60
21	J,	—	34	—	—	34
36	„	—	89	—	—	89
41	»»	—	95	—	—	95
51	„	—	97	—	—	97
61	,,	—	19	—	—	19
67	J,	—	30	—	~	30
6	„	—	34	—		34
3	Self purple	61	5	12	2	83
5	„	57	19	14	5	95
11	„	54	17	22	5	98
19	f.	53	16	27	6	102
26	,,	58	19	15	10	102
28	.,	54	15	19	6	94
29	»»	65	11	17	3	99
34	.,	58	16	14	8	96
35	)•	•62	13	19	5	99
46	f*	46	22	22	5	9.->
50	„	34	8	16	4	<!2
54	„	37	16	14