NOTEMBEn 15, 1907] 



SCIENCE 



655 



forgive me for taking a botanical illustra- 

 tion. I have no doubt it will not be long 

 before eases in animals are found. 



In the sweet pea, then, we know ex- 

 perimentally about eleven distinct allelo- 

 morphie pairs. The actual number is, of 

 course, much greater, but eleven have been 

 critically demonstrated. 



Of these characters some are concerned 

 with the production of color, others with 

 the determination of form. The composi- 

 tion of the F^ families shows that several 

 of these allelomorphs are not distributed 

 independently among the gametes, but that 

 certain combinations of characters occur 

 with greater frequency than others. The 

 first of these couplings to be made out was 

 that between the normal or long pollen 

 shape and the factor which determines 

 hlue color. In the absence of the long 

 pollen factor, the pollen is round. In the 

 absence of the factor for blue, the flower 

 color is red. The coupling here is such 

 that the F^ numbers instead of being 9 

 blue-long + 3 blue-round + 3 red-long +• 

 1 red-round, are 41 : 7 : 7 : 9, or very near- 

 ly so. 



This system would be produced by the 

 following gametic series : 7 blue-long + 1 

 blue-round -|- 1 red-long -|- 7 red-round. 



It is not possible to decide strictly wheth- 

 er the series is 7, 1, 1, 7, or 8, 1, 1, 8, and, 

 of course, the dichotomies which produce 

 the one or the other of these systems must 

 be entirely different, but the total of the 

 series is either 16 or 16 -{- 2. 



Now the other two instances of partial 

 coupling show that the association is there 

 in groups of either 32 or 32 + 2. In the 

 first case th'e blue factor and the pollen 

 shape are again concerned, but their proper 

 system of coupling is disturbed by the pres- 

 ence of another element, that which gov- 

 erns the shape of the flower. 



The three pairs of characters are then : 



Dommant Recessive 



1. Blue. No blue, viz., red. 



2. Pollen long. Pollen round. 



3. Standard upright, hav- Standard hooded, with- 

 ing central notch. out a central notch. 



Now, experiment has shown two things. 

 First, that in these families there is a total 

 and complete coupling of blue and hood. 

 In other words, all gametes destitute of the 

 upright standard factor have the blue fac- 

 tor, while all gametes bearing the upright 

 standard are destitute of the blue factor. 

 Consequently, there are in such families 

 three types of plants, distinguishable by 

 the shape and color of their flowers : 



1. Blue — hooded standard. 



2. Blue — erect standard. 



3. Red — erect standard. 



Classes 1 and 3, are homozygous, but 2, 

 which in this curious instance happens to 

 be the wild type of sweet pea, is here al- 

 ways heterozygous, like the blue Anda- 

 lusian fowl. Consequently we meet the 

 paradoxical result that of the three types 

 produced in such a family the original wild 

 form is the one which does not breed true, 

 but continues to throw ofi! the other two 

 types. 



It is only by a stretch of language that 

 we can speak of the blue factor as coupled 

 with the hooded shape; for the hooded 

 shape is recessive, and thus may be re- 

 garded as the shape due to the removal of 

 the factor for upright standard. A more 

 strict way of describing the facts would be 

 to speak of erect standard and blue factor 

 as gametically alternative to each other. 

 It is thus possible that we may have eventu- 

 ally to extend the conception of allelomor- 

 phism to cases like this where two char- 

 acters, both dominant, due, that is to say, 

 to the presence of some factor, are alterna- 

 tive to each other in the constitution of the 

 gametes. 



To return now to the distribution of the 

 pollen characters in these families : the F. 



