4l6 Scientific Proceedings, Roijal Dublin Society. 



that is recessive to P and dominant to r, but tliis is unlikely. It would mean 

 a series lilce that of the horse colours. Assume Rr to be a new pair for the 

 present. The assumption can be dropped later if found inconsistent with the 

 facts. Then the provisional scheme for these four colours and the two left 

 behind in the first experiment becomes 



But we can go farther. The first experiment showed that agouti and chocolate 

 differ in only two characters. Therefore agouti contains B. The same 

 experiment showed also that cinnamon agouti and agouti differ in only one 

 character. Cinnamon agouti, tlierefore, also contains R. The second experi- 

 ment showed that blue differs from black and from silver-fawn each in one 

 character. Therefore blue and silver-fawn both contain j). We can now write 

 these six colours more fully : — 



At this stage it can be predicted that, when the complete set of colours is 

 worked out, there will be eight groups in all, viz., one for every combination 

 of six characters in groups of three. 



In Miss Durham's third experiment, agouti and blue gave 9 agoutis : 

 3 dilute agoutis (a new colour) : 3 blacks : 1 blue.^ Thus dilute agouti differs 

 from agouti, black, and blue, each in one factor. Its composition might be 

 found in several ways. The simplest is to write down the possible combina- 

 tions in groups of three of the six factors PQRpqr, and select that which fits 

 the case. There are eight groups, viz., PQFi,, pQR, PqR, PQr, Pqr, pQr, 

 pqR, and pq): Since dilute agouti differs from agouti in one factor, it must 

 contain two dominants and one recessive, and it must be found therefore in 

 the second, third, or fourth group. It cannot be pQR, since that represents 

 black. It cannot be PqR, since then it would differ from blue in three 

 characters. It can only be PQr. 



In Miss Durham's fourth experiment, the last of the eight groups was 



1 "Journal of Genetics," vol. i, No. 2, p. 177. 



