650 



Large Larch Saw-Fly. 



[dec, 



mentioned. Packard does not appear to have seen one. This 

 is no doubt on account of its rarity, N. erichsoni has been 

 shown by the present investigation to produce almost solely 

 by parthenogenesis, that is, the larvae hatch out from eggs 

 which have not been previously fertilised by males. This 

 phenomenon occurs in other species of saw-flies (Tenthredinidae), 

 for example, in the common Gooseberry or Currant Saw-fly, 

 N. ribesii. That this form of reproduction takes place was 

 proved not only by segregating the females immediately on their 

 emergence from the cocoons, but also by the fact that out of 

 300 cocoons which were specially observed only 2 males emerged 

 to 298 females. 



Description of Male. — The male is smaller than the female. 

 Its length is 9 mm. and expanse of wings 15 mm. (the female 

 is 11 mm. in length with a wing expanse of 20 to 21 mm.). The 

 terminal portion of the abdomen is broadly rounded. The 

 legs are paler in colour than those of the female, and only a 

 small portion of the distal extremities of the tibiae of the third 

 pair of legs is dark coloured. 



Life-history. — The female saw-flies began to emerge from the 

 cocoons on 27th April (at an average temperature of 55 0 F.),. 

 but the largest number emerged from 29th May to 3rd June. 

 At my request, Mr. Edwards, Forester on the Manchester Water- 

 works estate at Thirlmere, kindly observed the appearance of the 

 saw-flies in the plantations, and the first was recorded on 

 1st June, which agrees with the period of maximum emergencies 

 in my experiments. I observed females still omthe wing in the 

 Thirlmere district towards the end of July. From this it 

 might be inferred that there were two annual broods, such as 

 occur in certain other species of saw-flies, for example, in the 

 Pine Saw-fly Lophyrus pini, but I was not able to obtain two 

 broods in my experimental studies, even from the earliest 

 emerged batches. The emergence of the saw-flies from the 

 cocoons of the previous year under natural conditions extends 

 over a period of about eight weeks, and this fact explains the 

 frequent occurrence of both full-grown and recently hatched 

 larvae on the same tree, which, together with the presence of 

 the adult saw-flies over such c,n extended period, might be 

 taken as indicating that the species was double-brooded, in 

 the absence of evidence to the contrary. 



