666 



SCIENCE. 



[N. S. Vol. XXIV. No. 621. 



other earlier evolutionists) has given it a cer- 

 tain standing in the literature of bionomics. 



A student in my laboratory. Miss Lilian 

 Eamsay, in a general study of variation and 

 heredity in certain insects, has been able to 

 gather some definite data in regard to as- 

 sortative mating (the basic assumption of the 

 physiological isolation theory) in the case of 

 an insect species common on our university 

 campus. Unfortunately the earthquake of 

 last April interrupted her interesting observa- 

 tions. Her data are the following: 



The lady-bird beetle, Hippodamia conver- 

 gens, is an insect of much variability as to its 

 dorsal color pattern (dorsal surfaces of the 

 two wing-covers). The modal type shows a 

 red-brown ground bearing twelve small black 

 spots (six on each wing-cover). The varia- 

 tion in additions to or subtractions from these 

 normal twelve spots, and in their arrangement, 

 is so large, extending from a condition of 

 total absence of spots to that of the presence 

 of eighteen distinct spots, that we have been 

 able to describe, in words, some eighty-four 

 * aberrations ' or pattern^ types of this species. 

 As any of these variations may appear in con- 

 junction with any others in a single brood 

 (one clutch of eggs), it is certain that these 

 aberrations are all included within the spe- 

 cies. Is there now any assortative mating 

 among these variations? 



In March and April Miss Ramsay found 

 the beetles mating on the wild mustard on 

 which their food, the mustard aphis, swarms. 

 Collecting these actually mating couples, she 

 brought them into the laboratory, noted the 

 pattern types to see if assortative mating on 

 a basis of pattern — this being largely the basis 

 for specific separation in the genus — was oc- 

 curring, then put each couple into a breeding 

 jar to obtain the eggs and rear the young, to 

 ascertain if the various matings were similarly 

 fertile. The earthquake interrupted the col- 

 lecting and made impossible the determina- 

 tion of fertility. The actual matings of sixty 



* See Kellogg and Bell, ' Studies of Variation in 

 Insects,' Proc. Wash. Acad. 8ci., Vol. 6, pp. 203- 

 332, 1904, for a detailed account of the variation 

 of the species. 



couples as they occurred normally in nature 

 were as follows: 



Thirty cases of twelve-spotted Hippodamia 

 convergens with twelve-spotted H. convergens. 



Thirteen cases of twelve-spotted H. con- 

 vergens with spotless H. convergens. 



Five cases of spotless H. convergens with 

 spotless H. convergens. 



Two cases of twelve-spotted H. convergens 

 with two-spotted H. convergens. 



Two cases of one-spotted H. convergens with 

 spotless H. convergens. 



One case of twelve-spotted H. convergens 

 with ten-spotted H. convergens. 



One case of two-spotted H. convergens with 

 spotless H. convergens. 



One case of spotless H. convergens (female) 

 with Megilla vittigera (male), a distinct spe- 

 cies with three strong, longitudinal stripes 

 instead of spots. 



Also there were noted seven cases of the 

 mating of plain Coccinella californica with 

 plain C. californica, and one case of plain 

 C. californica with two-spotted C californica. 

 Coccinella californica also frequently mates 

 with Hippodamia convergens, as we have ob- 

 served at various times. 



As regards relative abundance of the vari- 

 ous variation types of Hippodamia and Coc- 

 cinella, the twelve-spotted and the spotless 

 Hippodamia individuals far outnumbered any 

 other types and the spotless Coccinellas were 

 much more numerous than spotted ones. 



The matings, therefore, seem obviously to 

 be wholly non-selective; they are chance ma- 

 tings, that is, follow the law of probability. 

 All the individuals, spotless, twelve-spot and 

 few-spot, mingle freely on the same mustard 

 plant, so that the matings might readily be 

 assortative. If not assortative, then the rela- 

 tive proportion of numbers of the different 

 types should determine the matings. And 

 this is what apparently really obtains. 



Vernon L. Kellogg. 



Stanford Univeesity, Calif. 



THE SALARIES OP PROFESSOBS."^ 



Despite the heavy burdens upon the cor- 

 * From the report of President Butler, of Co- 

 lumbia University. 



