542 



SCIENCE 



[N. S. Vol. JlXV. No. 640 



the laws governing its inheritance. The effort 

 has apparently been completely successful. 

 The case is one of exceeding interest, and 

 proves to be a complex case of Mendelian 

 inheritance. There is some probability that 

 it is a case of reversion to a very ancient type 

 related to the Indian tapir, which is somewhat 

 similarly colored. At first it was suspected 

 that the belt was due to a simple Mendelian 

 character. According to the testimony of 

 breeders, when two belted hogs produce black 

 progeny there is a distinct tendency for one 

 fourth of the litter to be black. Also, when a 

 belted animal is bred to a black, two cases are 

 evident: first, the litter may all be belted; 

 second, in some cases there is a distinct tend- 

 ency for half the litter to be belted and half 

 black. Thus far, we seem to be dealing with 

 a simple character pair. But all breeders 

 agree that black bred to black does not always 

 produce black. In fact, litters of this breed- 

 ing may be all belted. 



Professor Castle accounts for the agouti 

 color in guinea pigs by supposing a third inde- 

 pendent character which governs the arrange- 

 ment of the red pigment in the hair, so that 

 the black and red pigments together, under 

 the influence of this third character, produce 

 the agouti color. We can fully explain the 

 peculiarities of the inheritance of the belt 

 character in Hampshire swine by a somewhat 

 similar assumption. In this case, however, 

 this third character, which governs the distri- 

 bution of white and black over the body, is a 

 compound character of two factors. Repre- 

 senting the two factors by F and O, and the 

 two colors by W and B, the allelomorphic 

 formula of a pure belted individual is FF, 

 GG, WW, BB. The individuals lacking the 

 belt have white fore feet, so that W and B 

 are present throughout the breed. We may, 

 therefore, omit W and B in our formulae. If 

 either F or G is entirely absent the belt fails 

 to appear. Representing the absence of either 

 F or G hy 0, the following nine types are 

 possible, and doubtless occur in the breed: 



TABLE I. 



o. FFGG, producing gametes FO. 



6. FFOO, producing gametes FG, FO. 



c. FOGG, producing gametes FG, OG. 



d. FOGO, producing gametes FG, FO, OG, 00. 



e. FFOO, producing gametes FO. 



f. FOOO, producing gametes FO, 00. 



g. OOGG, producing gametes OG. 



h. OOGO, producing gametes OG, 00. 

 i. 0000, producing gametes 00. 



Types a, h, c and d are belted; the others 

 are not. 



Table II. shows the proportion of ' blacks ' 

 in the progeny of all the possible conjugations 

 of these nine types. 



Table II. may be divided into three sections, 

 as shown by the division lines. The first sec- 

 tion shows the results that follow when belted 

 hogs of all types are bred together. Of the 

 ten possible types of breeding in this division 

 five result in no blacks, four give 25 per cent, 

 blacks, and one 43-|- per cent. Breeders have 

 already recognized all these except the last, 

 which occurs in only one out of ten possible 

 cases. 



In the second section of the table are found 

 the results from breeding belted hogs on 

 blacks. Of the twenty cases seven give all 

 belts; eight give 50 per cent, blacks; two give 

 25 per cent, blacks; two 62-|- per cent; and 

 one gives 75 per cent, blacks. The results 

 which occur most commonly have both been 

 recognized by breeders, namely, cases giving 

 100 per cent, belts, and those giving 50 per 

 cent, blacks. The remaining cases occur so 

 seldom that they have not been recognized. 



The third section of the table displays the 

 results from breeding black on black. Of the 

 fifteen cases, eleven give blacks only, two give 

 50 per cent, blacks, one gives 25 per cent. 



