THE INttERITANCE OF BLACK, YELLOW AND TORTOISESHELL COAT-COLOUR IN .

BY RUTH C. BAZ~tBEI~ (M~s BISBEE), M.Sc., F.L.S. AND E. CATHERINE HERDMAN, M.Sc. (The University of Liverpool, 1926.)

THE problem of the inheritance of black, yellow and tortoiseshell coat-colour in cats is both complicated and interesting. Comparatively few careful~y controlled experiments have been carried out, but consider- able data have been collected from brceders, and certain facts have emerged quite clearly, h~ormally the cross black ~ • yellow 3 gives tortoiseshell ~~ and black 33: the reciprocal cross, yellow ~ x black 3, gives tortoiseshell 99 and yellow 33. Tortoiseshell malea occur only as very rare exceptions. These observed facts have been explained in many ditIerent ways. Doncaster (1904) suggested that yeUow is completely dominant over black in the male, incompletely dominant in the female, so giving tortoiseshell daughters and ycllow sons from the cross yellow ? x black 3. Little (1912) pointed out, however, that the reciprocal cross black ~ • yellow 3 gives black sons and tortoiseshell daughters, and that therefore dominance of yellow over black in the male is notan adequate explana- tion of the observed facts. He suggested that both yellow and black are sex-linked, a black o being BB, a black 3 B-, a yellow ~ YY, anda yellow 3 Y-. In the same year Doncaster (1912) concluded that yellow only is sex-linked, and that black is present in all gametes, yellow being completely dominant over all black in the male, incompletely dominant in the female. In 1913 Johannsen put forward a theory almost exactly like that offered by Doncaster in 1904, except that he looked upon black as being dominant to yellow in the male, whereas Doncaster supposed yeUow to be dominant to black. In 1916 Ibsen suggested that tortoise- shell is due to a separate factor, T, which can only act in the presence of black, B, causing ir to be restricted to spots with yellow areas between. He considered black, B, to be dominant to yellow, b, and suggested that normally T is closely linked with b. He supposed both pairs of allelo- morphs, Tt and Bb, to be sex-linked. Wright in 1918, discussing the chemistry of coat-colour, adopted Little's idea of sex-linkage of both 88 Coat-Colour in Cats black and yellow, bar evidently considered yellow to be simply absence of black. In 1919 Little modified bis earlier hypothesis to the extent of placing B, black, in aU X-chromosomes and making Y, yellow, and y, the absence of yellow, a separate pair of sex-linked allelomorphs. He suggested that one dose of yellow is coinpletely epistatic to one dose of black giving yellow in both sexes, but that one dose of yellow to two doses of black gives tortoiseshell. Apart from Doncaster's first suggestion (1904) and the similar one puf forward by Johannsen in 1913, any one of the above hypotheses will account satisfactorŸ for the observed normal facts of inheritance of black, yellow and tortoiseshell. There are, however, a number of exceptional facts of inheritance. A tortoiseshell male does occasionally appear, and somewhat more frequently black daughters ate produced from yellow sires. There ate also one of two eases on record of a yellow female having bred as a tortoiseshell. Many different explanations, based on one or other of the above hypotheses, have been puf forward to account for these excep- tional individuals, but no theory so lar offered has been able to bear the full weight of the recorded facts, and the whole problem of the inheritance of black, yellow and tortoiseshell remains unsettled. (A detailed dis- cussion of the rival theories is given by Bamber in Bibliographia Genetica, Vol. m. pp. 14-44.) Either yellow or black or both ate certainly sex-linked, but the ordinary facts of inheritance can be explained equally well by assuming: (1) that black and yellow are both sex-linked and ate allelomorphs, i.e. two positive factors whieh are alternative to one another; of (2) that both ate sex-linked but ate not allelomorphs, black being present in aH X-chromosomes, one dose of yellow being completely epistatic to one of black; of (3) that both ate sex-linked but are not allelomorphs, yellow being present in all X-chromosomes, one dose of black being completely epistatic to one of yellow; of (4) that yellow only is sex-linked and that black is present in all gametes (i.e. in ah autosome), one dose of yellow being completely epistatic to all black in the male, incompletely so in the female; of (5) that black only is sex-linked and that yellow is present in all RU~H C. BAMBER AND E. CATHERINE I-]ERDMAN 89 gametes, one dose of black being completely epistatic to all yellow in the male, incompletely so in the female. Two of the above possibilities (3 and 5) ate, however, usually dis- counted because at least three matings between yellow cats have been definitely recorded as giving black or tortoiseshell in the offspring (Don- caster, 1913; Whiting, 1918), whereas there is only one doubtful record of two blacks ever having given tortoiseshell in the offspring (Bonhote, 1915). Notwithstanding the contticting evidence ir seems certain that, whatever may be the exact relationship between black and yellow, either there is a difference in dominance in the two sexes or both colours ate sex-linked. The observed results of carefully controlled experiments are few. Little (1912) and Whiting (1915, 1918, 1919) have both recorded results from their own experiments but these involve comparatively few matings. Doncaster (1913) has given the fullest account of the inherit- ance of black, yellow and tortoiseshell but bis data were collected from breeders for the Fancy and were not the results of carefully controlled scientifie experiments. Bonhote (1915) also gives many facts, but again the majority were obtained from breeders, and most of these facts had already been given to Doncaster and were included in his 1913 paper; they ate, therefore, no additional evidence. Thus there is nota very firm foundation of fact on which to erect such a weighty superstructu~e of hypothesis. In the hope of throwing some additional light on the subject we have been carrying out expe¡ with cats during the past four years. The experiments have been chiefly genetical but a little physiological work has also been included. One of us suggested, Bamber 1922, that, ir there is a difference of dominance between black and yellow in the two sexes, ir might be possible to alter the eolour of a male by turning his physiology towards the Ÿ condition. In an attempt to do this, in 1923, three newly born yeUow male were castrated, and, ffom the time they were about four months old, two of these were red with ovarian extraer for a period of six months. The results were entirely negative. We intended to perŸ the same experiments with black males, but, before this was attempted, our genetical work practically proved that there is no difference in dominance in the two sexes, and our physiological work was therefore discontinued. In reporting ou~ breeding experiments, it is important to state clearly that our animals have been kept under conditions which preclude 90 Coat-Colour in Cats the possibility of any inaccuracy in our records. The females have Iived and bred in separate cat-houses with enclosed wired-in grass runs, and have notat any time been allowed liberty. The males have been free to roam the neighbourhood 1, but have each been confined in a separate pen with a single female at the times of mating. Our results, as recorded in this paper, ate, therefore, entirely reliable. Many different crosses have been made, and in each case there have been several litters. The total results of these experiments ate as fol]ows : (a) Black c7 • Black 9--5 Black 8c~; 8 Black 99. (b) Black ~ • Yellow 9--3 Yellow ~~; 3 Tortoiseshell 99. (c) Black c? • Tortoiseshell 9--11 Black ~c~; 8 Yellow c~c7; 9 Black 99; 8 Tortoiseshell 99. (d) Yellow c7 • Yellow 9--see (g) below. (e) Yellow c7 x Black 9--7 Black 8c~; 2 Tortoiseshell ~~2. (f) Yellow c7 • Tortoiseshell 9--1 Black ~; 5 Yellow ~~; 4 Yellow ~; 1 Tortoiseshell 9 ; 1 Anomalous u 9 (4 d) 2 (g) Yellow ~ • Anomalous Yellow 9 (4 d)3--3 Yellow r 2 Yellow 99; 1 Anomalous Yellow ~ (Tortoiseshell?) ~ These results ate given in detail in Table I. Ir seems advisable to do this both because the amount of data from controlled experiments is smaU and because, so often, in condensing results, facts disappear which would be of value to other workers approaching the problem from a different view-point. Also, in such a controversial subject, any fact may prove to have an unsuspeeted signi¡ Reference numbers ate given in the table Ÿ all an~mals which have been used for breeding, so that the interrelationship of our stock can be easily followed. Arable numerals denote individual cats; Roman numerals denote matings.

1 There is, however, no question as to the identity of the tom-cats. They were treated ss personal friends and, as such, were readfly recognisable from other cats which visited the cattery from time to time. z The ~ was yeUow with s minute amount of black on the bsck of her right l¡ loor, snd n¡ therefore be caUed tortoiseshell. She is here elassified as yellow because her breeding behaviour suggests thst her yellow and black do not segregate as they do in a normal tortoisesheU; sbe sppe to be homozygous for yellow in addition to having the smaU amount of black. Her son, the "anomMous yeUow" male had about twiee as much black as bis mother, but bis type of colouration is so obviously the same as hers that ir would be absurd to clsssffy them differently. The whole subject is discussed in detail on p. 92. RUTH C. BAMBER AND E. CATHERINE HERDM~N 91 TABLE I. Tor- Anota. NO. of Black Yellow toise- yeUow Mating Nature of mating I. r 3 Black • ~ 2 Black 1 2 ...... II. • 3 1 ...... IlI. 1 a Black (ex XV) x 9 2 Black 1 3 ...... IV. d~ 1 a Black (ex XV) • 9 16 b Black (ex II} 2 ...... V. 1 a Black (ex XV) • ~ 2 a Yellow (ex XVIII) 3 2 VI. ~~ • ~~ ..... 1 VIL 1 a Black (ex XV) x ~ 4 Tortoiseshell 2 .... 1 VIII. 1 . 2 1 IX. X 1 1 X. 1 a Black (ex XV) x~ 5 Tortoiseshell XI. 3 1 1 ..... XII. XIII. 7 Black • 9 5 Tortoiseshell 32 1 1 XIV. 13 Black • 5 Tor~oiseshell 2 2 1 1 XV. 1 Yellow • ~ 2 Blaek 5 ...... XVI. ,, • ,, 2 .... 1 XVII. ~ 12 Yellow • 16 b Black (ex II) ..... 1 XVIII. ~7 1 Yellow x ~ 5 Tortoiseshell 3 1 .... XIX. 1 Yellow • 4 Tortoiseshell i 1 1" 1" XX. 12 YeUow x$ 4 Tortoiseshell 1 2 i 8 XXI. r 1 Yellow • 9 4 d Anomalous 1 1 i YeUow (ex XIX) XX_II. ~ 9 Yellow x ~ 4 d Anomalous 2 1 .... Yellow (ex XIX) * This kitten was recorded asa 9, but as ir died young and w~ not dissected its sex is not absolutely certain. On the whole our results ate in harmony with the flndings of other workers, but some new facts have come to light which gire rise to en- tirely new points of interest. These facts ate: (1) A yellow ~ (4 d) has appeared with a minute amount of black- spotting on the back of one foot and this black seeins not to have segregated from yellow in her germ ceUs--it was apparently transmitted together with yellow to one of her sons. (2) Twenty-six yellow cats have been examined and, with one possible exception, all have been found to have a few scattered black hairs. Ah equally careful exam~nation of forty-three black cats revealed only one single yellow hair. The yellow ? (4 d) was at f~st thought to be ah ordinary yellow kitten, but when she was about four months old three minute black spots were discovered on the back of her right hind foot. She was then, of course, recorded asa tortoisesheU. Naturally she reminded one of the two yellow females, recorded by Doncaster (1913) and Whiting (1918), which bred as tortoiseshells, and she seemed to support the suggestion made by Whiting (1915) that tortoisesheU females may vary f~om solid black to solid yeUow. Whiting (1915) and Doncaster (1913) have both recorded Maltese ("blue" = dilute black) eats with only a few creara 92 Coat-Colour in Cats hairs. Whiting states that his tortoiseshell females could be arranged in a series leading from one predominantly yellow to one almost pure Maltese, and our almost entirely yellow female, together with the two entirely yellow ones referred to above, which bred as tortoiseshells, seemed to complete the yellow end of the series. To aecount for such a series Whiting (1915) put forward the suggestion tha~, in addition to the colour factors involved in the production of tortoiseshell, there ate also modifying factors governing the relative amounts of the two colours and so producing continuous variation from yellow to black. It was, therefore, of interest to test the mode of inheritance of the yellow and black in our anomalous yellow female (4 d). She herself was the offspring of a yellow male (1) anda dark mingled tortoiseshell female (4). When mature she was mated with her own father (4 d • 1) and produeed a yellow male, a yellow female and an "anomalous yellow" male. This male was extremely interesting in that Iris type of marking was clearly the same as his mother's. On casual observation he might have been mistaken for a yellow. He had, however, more black than bis mother and its distribution was different: he had a distinet spot over the left eye, a very small spot above each ear, a scarcely visible one on the right shoulder, a small one on the left side of the tafl about hall way down, and a minute one on the right side of the tafl nearer the tip, in addition to numerous scattered black hairs throughout the yellow parts of his eoat. This small amount of black-spotting seems ahnost certainly to have been inherited from bis mother. But he was predominantly yellow, and visible yellow is not normally transmitted from a yellow male to his son, so that ir is almost certain that he received Iris yellow also from bis mother. But blaek and yellow segregate in a normal tortoiseshell female. It looks as though our female (4 d) were really a yellow, but with some of her yellow-carrying gametes "contaminated" with blaek 1. As i As there is apparently no rule of inheritance in cats without its exception, it is doubt- less possible that in this case the yellow father (1) did gire yellow to hiz son and that the hlack only carne from the mother, in which case she would be a tor~iseshell carrying some unusual modifier of the amount of black, which was also transmitted to her son. Ir is against the laws of probability, however, that the male (1) should transmit yeUow abnor- mally to a son just on that one particular occasion when he was mated with ah unusual female---all bis other known offspring were normal in this respect. He was, however, the father of 4 das well as of her abnormal son, and this does introduce an element of doubt into the question he may have been responsible for the unusual colouration of 4d and in turn for that of her son, in spite of the fact that MI his other progeny were normal and that he hŸ was, in appearance, a peffectly normal yellow: but it seems improbable. By far the most probable explanation of the whole matter seems to us to be that given above, viz. that 4 d is abnormal in colouration, i.e. is an anomalous yellow and nota tortoiseshell, and that she has handed on her abnormality in toro to her son. RUTH C. BAMBER AND E. CATHERINE HERDMAN 93 such she has, of course, no bearing on Whiting's series, if his animals were all genetically true tortoiseshells. Later she was mated to a different yellow male and produced two yellow males and one yellow female, all without any black-spotting. Unfortunately soon after this she died. Her "anomalous yellow" son had been drowned and preserved in forma- lin during our absence and bis father had strayed and been lost during the same period. That part of our investigation, therefore, carne to an untimely end and left us with only a tantalising suggestion. After the discovery of the black spots on ~ 4 d we naturally examined all our yellow cats very carefully, but none had the least suggestion of a black spot anywhere. Later one of our yellow kittens became infested with fleas and, when going over ir very earefully with a small tooth comb, we found three of four black hairs. No ordinary examination would have revealed this small amount of black and consequently we began to examine every available yellow cat with especial care. We have examined eats from our own stock, from the Cats' Home here in Liver- pool, from different parts of England and from the Isle of Man, aud the interesting faet has come to light that apparently all yellow cats have a few blaek hairs. Altogether we have examined twenty-six individuals --males, females and neuters (castrated males)Iof all ages, and, in all except one newly born kitten, we have been able to find some black ha[rs. In four the black ha[rs were distinctly ticked as in tabby pattern, and presumably these animals were carrying the factor for agouti (ticking). There is no question of these yellow cats being tortoiseshells near the yellow end of the series; they ate ordinary yellows, and in those cases where we have bred from them they have shown no suggestion of segre- gation between black and yellow. It is doubtful whether a pure yeUow cat existsl: all are slightly "contaminated" with black. We have also examined forty-three black cats from the same distriets as the yellows but have succeeded in finding only one solitary yellow ha[r; this was on a female. These findings lead to several points of interest. In the first place they very strongly suggest that both black and yellow ate sex-linked. As was pointed out on p. 88, the ordinary facts of inhe¡ can be explained only by assumivg either that both black and yellow ate sex-linked, of that one is sex-hnked and the other not, the sex-linked colour being completely dominant in the male, incompletely

1 We cannot attach much importance to the one newly born kitten in which we fafled to find any black. Often in very young kittens it is extremely difficult to find the bl ten minutes' search with a hand lens may reveal only two or three black hairs. 94 Coat-Colour in Cats dominant in the female. The fact that in yellow cats, males and females alike show a few black hairs practically proves that there is no difference in dominance between the sexes, and goes lar to establish the sex- linkage of both colours. Ir might be suggested that we ate dealing here with a different black from that of an ordinary . Ordinary black in cats is recessive to the ticking which gives tabby. Such recessive black occurs in several other anima]s, while in some there is a black which is dominant to agouti. In the rabbit both dominant and recessive black ate known (Punnett, 1912) and there is good reason to believe that a dominant black exists in cats also. We have not yet had an op- portunity of testing the matter, but from Tjebbes' (1924) work on Siamese cats ir seems almost certain that such a dominant black does exist. He crossed a Siamese female with a tabby male and got aU blacks in the F i generation, and these crossed amongst themselves gave some tabbies in the F2 generation. This F1 black is clearly not the black of ordinary cats, for that is recessive to tabby. Nothing is known of this dominant black, nor of its place in the general scheme of inheritance of coat-colour in cats i, but its existence seems fairly certain and ir is necessary, therefore, to consider the possibility that this is the black which shows up in yellow cats. Ir this should prove to be the case, then the scattered black hairs would not necessarily have any bearing on the problem of sex-linkage of ordinary black and yellow. But, as already mentioned, in four yellow cats examined by us, the black hairs were ticked, whereas the dominant black of Tjebbes' cats did not show ticking. Ir is, therefore, practically certain that these scattered black hairs ate not dominant black, and, unless some entirely new facts come to light, ir is reasonable to suppose that they ate the same recessive black as that of ordinary cats. Ir this be true, it seems certain that there is no sex- difference in dominance between ordinary black and yellow, and ir follows, therefore, that both colours must be sex-linked. This still leaves three alternatives, namely, that black and yellow ate a pair of allelomorphs 2, of that black is epistatic to yellow, or that yellow is epistatic to black. On tliese points, however, we have no real evidence. There also arises from our results the larger question of the possi- bility of the splitting of a factor. Although in the present state of our 1 Since going to press we have seen a second paper by Tjebbes, 1926, drawing atten- tion to this dominant black in caos and suggesting a possible connection with the problem of the tortoiseshell tom-cat. This involves the whole question of the validity of the Presence.and-Absence theory. For a discussion of this subject see Bateson (1926). RUTH C. BAMBER AN:D E. CATHERINE HERDMAN 95 knowledge we cannot pretend to give an exact scheme of the relationship between black and yellow, it does seem to us that in any case our results can best be explained on the supposition of what Bateson (1926) calls the "fractionation" of a factor. Presumably in one of the gametes which went to produce 4 d, something happened which caused a small amount of black-spotting to appear in her coat, in spite of the fact that she was evidently what is usually understood by a yellow. Ir may hace been that the greater part of the factor for black was lost, allowing a hypo- static yellow to become visible; or ir may have been the partial loss of the factor for yellow, allowing hypostatic blaek to appear. In either case, partial non-disjunction may have combined with fractionation, causing a part of the factor for one colour to adhere to a chromosome which would normally have determined the other colour. Ir black and yellow be a pair of allelomorphs such partial non-disjunction is, of course, the only one of the above alternatives which could be applied to our resulta. What- ever be the exact relationsl¡ between black and yellow, however, the fact remains that a quantitative difference in one of them has arisen during the course of our experiments, in 4 d and her son, and has evidently a¡ also in the past, giving the very small amount of black which is found in all homozygous yellow cats. Ir seems to us that these quantita- tive differences strongly suggest the splitting or fractionation of a factor. Doubtless many workers would prefer to explain our results by linked multiple factors for black, some of which, by crossing-over, might come to lŸ on the same chromosome as the factor for yellow (or vice versa). This line of argument easily leads to absurdity in those cases where the se¡ of known forros passes fxom one extreme to the other through ah unbroken chain of individuals separated from each other only by minute quantitative differences of a single character 1. Again, it might be suggested that the small amount of black in ordinary yellow cats and the greater amount in 4 d and her son ate due to restriction factors affecting the black, or extension factors af~ecting the yellow. This is a possible explanation of the facts under consideration. Unless, however, there is almost perfect lin~age between these modifying factors and the colour factors ~, intermediate grades, such as 4 d and her son, showing both black and yellow and yet not breeding as true tortoiseshells, should be fairly common, whereas in reality they ate extremely rare.

1 E.g. A complete series exists between black and dominant white in cats. This subject will be discussed in a separate paper. 2 To suppose almost peffect linkage m~kes this only a speciMised case of multiple factors. 96 Coat-Colour in Cats Ir seems to us more reasonable to accept "yeUow," 4 d, and blaek as a series of multiple allelomorphs, which ate only different degrees of a single character, and to suppose that these different degrees have arisen by the splitting of a single factor. The only diffieulty in the way of such ah explanation is the deeply rooted but purely hypothetieal con- ception of an indivisible gene.

SUMMARY. (1) The normal facts of inheritance of black, yellow and tortoiseshell can be explained only by supposing either that both colours ate sex- linked, or that one only is sex-linked and eompletely dominant to the other in the male, ineompletely dominant in the female. (2) Black hairs have been found in aH yellow cats of either sex examined by us, and this practieally disproves any sex-ditterence in the dominance of black and yellow. There remaius, therefore, as the only alternative, the sex-linkage of both colours. (3) In the course of our investigations a yellow remate appeared, showing and trausmitting a very small amount of black-spotting whieh did not segregate from the yellow. We subsequently found that all yellow eats have a few black hairs. These faets suggest that in the past and again during our expe¡ there has been fractionation of a factor.

(Our thanks ate due to the Royal Society of London for grants which have enabled us to carry out the investigations reported in this paper.)

LITERATURE.

BA~BER, R. 0. (1922). "The male tortoiseshell cat." Journ. Genetic~, xII, no. 2, p. 214, footnote. -- (1927). "Genefics of Domestic Ca BibliograThia Gcndica, m, pp. 14-44. BAT~.so~, W. (1926). "Segregation." Journ. Genetics, xw, no. 2, pp, 201-235. BOSHOTE, J. L. (1915). Vigour and Heredity, West Newman and Co., London, pp. 55-60. DONCASTER, L. (1904). "'On the inheritance of tortoiseshell and related colours in cats." Pro~. Camb. Phil. Soc., pp. 35-38. . (1912). "Sex-l~mited inhe¡ in cats." Science, N.S. x~~vi, p. 144. (1913). "On sex-limited inhe¡ in cats, etc." Journ. Genetic~, rrr, no. i, pp. 15-19. IBs~~, H. L. (1916). "TricoIor inhem'tance. No. 3. Tortoiseshetl Cats." Genaics, I, pp. 377-386. RUTH C. BAMBER AND E. CATHERINE HERDMAN 97

JOKANNSEN, W. (1913). Elemente der exakten Erblich]ceitslehre. Gustav Fischer, Jen&, 2 te Ausg., pp. 609-612. LIT~LE, C. C. (1912). "Pre|imiuary note on the occurrence of a sex-limited character in eats." Science, N.S. xxxv, pp. 784-785. -- (1919). "Colour inheritance in cats, with speeial reference to the colours blaek, yellow and tortoiseshell." Jour~. Gene~ics, v~I, pp. 279-290. PUNNE~rr, 1~. C. (1912). "Inheritance of coat-colour in rabbŸ Journ. Genetics, ~, no. 3, pp. 225-229. TJEnBES, K. (1924). "Crosses with Siarnese cats." Journ. Genetics, xIv, no. 3, pp. 355-366, especially Table B. TJEB~ES, K. and W~XEDT, C. (1926). "Dominant blaek in eats and its bearing on the question of the tortoiseshell males." Journ. Genetics, xv~, no. 2, pp. 207-209. Wm~No, P. W. (1915). "The tortoiseshell cat." Am. Naf., XLIX, pp. 518--520. -- (1918). "Inheritance of coat-colour in cats." Jourm of Exp. Zool., xxv, pp. 544- 549. ----(1919). "Inheritanee of white-spotting and other eolour eharacters in eats." Aro. Nat., Ln~, p. 474, WRmHT, S. (1918). "Colour inheritanee in Mammals, No. 10. The Cat." Journ. of Heredity, Ix, no. 3, pp. 139-144.

Journ. of Gen. xvlt~ ";