188-1 J 147 



of Kowalevski mentioned in my former communication. The earliest 

 eggs were laid on or about the 15th June. On the 28th, I noticed the 

 first appearance of the eye-spots, and the first hatching took place on 

 2nd July. My note on 30th June runs as follows : " The eyespota 

 from their earliest appearance occupy the same position relatively to 

 the sharp end of the polar oval as they do in these advanced embryos 

 (and which is their position up till hatching) : consequently the aspect 

 and orientation of the dorsal and ventral surfaces is constantly the 

 same." That is, unless the embryo makes, more than four days before 

 hatching, that revolution in the shell, asserted by Kowalevski for the 

 Lepidopterous embryo in general, and which would necessarily bring its 

 head from the one side of the shell to the other. The presumption, 

 then, would be that the embryo of B. cratcegata gets into the loop- 

 form by such a ventral incurvature and forward growth of the tail-end 

 as we have seen already in ZarcBci, and as is described by Huxley in 

 Astacus. 



Fi^om the cocoons made last year by parthenogenetic larvae (and 

 of which I had 26 remaining over winter), I had this year, in the 

 middle of April and beginning of May, three flies which were all 

 females, and of which the first two (excluded in April), laid eggs 

 abundantly, from which again I bred doubly parthenogenetic larvae, that 

 yielded me in June some 32 cocoons. Why I had no more than 3 

 flies from 26 cocoons may have been probably owing to the larvae pe- 

 rishing in the others from being kept in too dry and warm a situation 

 during winter. In Nematus rihesii the parthenogenetically-bred flies 

 being all males, agamic reproduction in the case of that saw-fly is 

 brought to a speedy termination in the second generation. The case 

 is, from what appears so far, very different with Zarcea, which may 

 possibly be capable of continuing the species agamically for an 

 indefinite time. This is so much the more likely as •the males of this 

 saw-fly appear to be very rare. I have only met with one hitherto, 

 excluded 8th June, out of 181 flies (and nymphs), 173 of which were 

 bred from larvae beaten out of snowberry in 1882. This male paii-ed 

 with a female much larger than itself immediately afterwards, and I 

 have at present a few cocoons the result of this union. 



Ichneumons, apparently of two species, made their appearance as 

 usual, on the average a good deal later than the saw-flies, so that the 

 larvae of the latter might be grown enough to receive their eggs, a 

 Zarcea larva nourishing only one ichneumon by which it is entirely 

 ' consumed. Besides the Hymenopterous parasites, I found also one 

 Dipterous cocoon, which, however, has not excluded any fly hitherto. 



