COLOUR IXHERITAXCE IN PIGEOXS. 617 



Grizzles tlian one would expect. So that here again, althoiigli 

 the numbers are small, the Mendelian proportion seems to be 

 npset by a slight tendency in a definite dii'ection. 



The occurrence of a pure Blue (no. 2) in Exp. 181 is the onbj 

 instance throughout the whole of our matings in which a colour, 

 other than one which was expected, has appeai-ed. Possibly the 

 male parent of this mating (1326), although unlike his brother 

 (1354) in extei'ual appeaiance, had in reality the same gametic 

 formula. Such an occurrence would not be without precedent, as 

 Mr. Staples-Browne instanced a homozygous Blue which showed 

 a certain amount of white, and this ought to have indicated a 

 heterozygous I)ird. The difierence between these two birds (i. e. 

 presence of white) is of the same nature, and in support of this 

 suggestion the result of the mating — 7. 2. 1— seems to be following 

 on the same lines as Exp. 182. 



There is one other possible suggestion for this abnormal result, 

 and this is that the White or Bed character may not have been 

 inherited with the Grizzle. 



The full formula of the parents on this basis is : — 



$ BBcd Gg [Ww]— cS BBcd Gg [RW]. 



ISTow if the Grizzle has been independently inherited, we might 

 get BBcdgg [RW, WW, wR, Ww] as the formula of one of 

 the offspring, and the last character [RW etc.] might not in the 

 absence of G be able to show itself, and this would give us what 

 we got, namely a pure Blue. The result of the mating of this 

 self-coloured bird (Exp. 192) throws no light on this, for if, as is 

 quite possible, her mate was a homozygous Grizzle, no selfs would 

 appear. 



It must be understood that these are mei-ely possible explan- 

 ations, of which we are inclined to favour the first, but at present 

 we have no definite proof in support of either. 



Eor the rest, we claim that our hypothesis is so closely borne 

 out by the facts that it may be accepted till further work 

 confirms or disproves it : and until we are clearly able to 

 differentiate between those characters which follow the Law of 

 Mendel and those which are apparently governed by other laws. 



In this paper we are only touching on the fringe of colour inheri- 

 tance in Pigeons, as there still remains the question of Black, 

 Dun, Red, Yellow, and White inheritance, on which we are at 

 present continuing our reseai-ches. Our work, however, empha- 

 sizes the fact that there are three important pi'oblems which the 

 Mendelian hypothesis fails to meet : — 



(i) The differences of shades in the same colour ; 

 (ii) The predominance of one sex in certain colours * ; 

 (iii) The gradual increase of the white in Grizzles and Mealies 

 in successive generations ; 



* This most interesting question has not been dealt with in the present paper, 

 as we have not yet fully investiuated the results ; but we may mention that a large 

 proportion of the White Grizzles are $ 's, and in the Light Mealies by far the 

 larger number are c? 's ; we have also bred a certain number of Cream Mealies, and 

 these have all been ^ 's. 



