Jahuart 19, 1906.] 



SCIENCE. 



113 



namely, (1) by pairing the yellow mice with 

 each other, and (2) by crossing them back 

 with pure grays, blacks or browns. In either 

 case, pure or homozygous F„ yellow mice (or 

 in Professor Morgan's view those that con- 

 tain only ' latent ' as opposed to ' free ' gray) 

 should give only yellow offspring, owing to the 

 uniform dominance of yellow, while mixed or 

 heterozygous yellows (yellow mixed with 

 ' free ' gray) should produce grays or other 

 colors as well as yellows. Cuenot says : ' J'ai 

 essaye par I'une et 1' autre methodes un nombre 

 considerable (81) de Souris jaunes * * *.' 

 ' Or, a mon grand etonnement, je n'en ai pas 

 trouve une seule (i. e., homozygote) .' Pro- 

 fessor Morgan considers this statement, which 

 embodies the principal result of Cuenot's ex- 

 periments, as ' somewhat ambiguous,' appar- 

 ently for the reason that Cuenot does not in 

 this passage actually use the words that all 

 the yellow mice produce offspring of other 

 colors as well as yellows ; but it must be obvi- 

 ous that only such a result could justify his 

 statement, and if this be not Cuenot's mean- 

 ing I am unable to discover any meaning in 

 his paper. In point of fact, however, he states 

 specifically on a preceding page (cxxvii) that 

 the cross between a yellow mouse and a pure- 

 bred one of a different color always gives off- 

 spring of this color (gray, black or brown) in 

 addition to yellows, the numbers being stated, 

 in the case of the yellow-gray cross, to be 

 equal. 



Now, according to Professor Morgan's as- 

 sumption there should on his own showing 

 be two classes of yellow mice in F.., of which 

 "the first group OY(CG) (i. e., those con- 

 taining 'latent' gray) will breed true, the 

 other group CY(CG)(CY)CG (containing 

 ' free ' gray) will split up in each successive 

 generation according to the Mendelian for- 

 mula." Such a behavior of the F^ yellow 

 mice was precisely Cuenot's expectation, but 

 ' to his great astonishment ' it was contradict- 

 ed by the results. Professor Morgan, never- 

 theless, insists that the case of the yellow mice 

 is precisely similar to that of extracted gray 

 dominants (both being ' contaminated ' by the 

 recessive character in the latent condition) 

 though, as Cuenot was the first to show, the 



latter breed true save for the rare appearance 

 of a different color, such as black, probably 

 derived from the latent color of the original 

 albino used. If the two cases do not differ, 

 why was so experienced an observer as Cuenot 

 astonished at his results, and why did he go 

 90 far out of his way to construct the special 

 hypothesis of selective fertilization to explain 

 the behavior of the yellow mice as distin- 

 guished from those of other colors? 



The difficulty with Professor Morgan's ex- 

 planation is that it proves too much, for it 

 explains the special peculiarities of the yellow 

 mice out of existence (!). My criticism is 

 not directed against Professor Morgan's gen- 

 eral assumption, but I think that it entirely 

 fails, as far as he develops it, to account for 

 the peculiarities of the yellow mice, and that 

 it leaves Cuenot's hypothesis, which it is sup- 

 posed to obviate, exactly where it stood before. 

 E. B. Wilson. 



THE LOGICAL BASIS OF THE SANITARY POLICY OF 

 MOSQUITO REDUCTION. 



The excellent address of Sir Eonald Ross 

 under the above title published in the Decem- 

 ber 1 niimber of Science, states the ^general 

 rules regarding mosquito distribution with 

 great accuracy; but it applies only to certain 

 species, including the Anopheles so far as 

 known to me and to Stegomyia fasciata. It 

 does not apply in the least to such forms as 

 Culex cantator and C. sollicitans. None of 

 the suggestions as to erratic flights that prac- 

 tically restrict the distance traveled influence 

 these species, which are truly migratory and 

 are guided by some motive other than finding 

 food or a place to breed. In fact, as I have 

 shown, these migrants never propagate their 

 kind and where they are to be dealt with, all 

 of the carefully reasoned mathematical deduc- 

 tions fall. The matter is of great practical 

 importance in New Jersey where communities 

 within whose boundaries not a mosquito 

 breeds, nevertheless, sometimes find life a 

 burden because of the insects. Local work in 

 such cases is worse -than useless. When we 

 find the dominant mosquitoes in the Orange 

 Mountains to be species whose nearest breed- 

 ing place is on Staten Island, time and money 



