MICHIGAN ACADEMY OF SCIENCE. 



47 



at the time of writing, and are entirely free from larvae. Further evidence 

 that the larvae subsequently found upon these plants were not in the egg 

 stage when the plants were brought in, is found in the fact that the larvae 

 on plants on which females were first placed on September 16 were notice- 

 ably smaller on a given date than those on plants on which females were 

 placed on September 5. Had they been in the egg stage at the same time, 

 that is, when the plants were introduced from outside, these larvae should 

 have been of equal size. I think it safe to conclude, therefore, that the 

 imported plants bore no thrips in any stage. 



Virgin females for the experiments were obtained by rearing large larvae 

 or pupae in isolation. When the adult emerged from the pupa, it was 

 placed, under cover, on one of the thrips-free mullein plants. In some 

 cases, only one individual was placed on a plant, and in these cases the 

 sex of the thrips was not determined. In other cases, the sex was deter- 

 mined by mounting the live insect in a drop of water under a supported 

 cover-glass, and examining it with a microscope. The males possess, on 

 next to the last abdominal segment, a pair of stout, dull spines, which are 

 wanting in the females. When the sex was determined in this way, several 

 females were put together on one plant. 



In at least six of these experiments larvae have appeared. The data from 

 these six experiments are given in Table 1. 



TABLE 1. 

 Showing details of experiments to test parthenogenesis of Anthothrips verbasci. 



There can be no doubt, I believe, that the larvae appearing in these ex- 

 periments were produced partheiiogenetically. Anthothn'ps verbasci, how- 

 ever, gives more evidence of sexual reproduction in nature than almost 

 any other species in the region around Ann Arbor. There are alwaj's nu- 

 merous males. Large numbers of spermatozoa are produced, as shoAMi by 

 sections of full grown larvae. There appears to be a lagging chromosome, 

 suggesting a dependence of sex upon fertilization. Furthermore, copula- 

 tion occurs frequently. In fact, it is usually necessary only to drive a male 

 and a female toward one another to induce mating. 



The discovery of parthenogensis in such a species greatly reduces the 

 probability that sexual reproduction is of common occurrence in the Thy- 

 sanoptera as a group. If Anthothri'ps verbasci is parthenogenetic, it is not 

 safe to conclude, without much better evidence than the abundance of 

 males and the occurrence of copulation, that any species is sexual. Whether 

 sexual reproduction ma}' occur only at certain seasons, or simultaneously 

 with parthenogenesis, is a question upon which the experiments here re- 

 ported have no bearing. 



