656 



Large Larch Saw- Fly. 



[dec, 



When the larvae were in the earlier stages it was found that 

 the three species of Tits — the Great Tit, Cole Tit, and Blue Tit — 

 fed upon them to a considerable extent. They were also 

 assisted by Chaffinches, which were found feeding on the full- 

 grown larvae. In addition to these birds, which perform no 

 little service, great destruction of the larvae was effected by 

 the Rooks, Jackdaws, and Starlings which were to be seen in 

 large flocks in and about the more seriously attacked planta- 

 tions. They not only fed upon the larvae on the trees but also 

 followed them on to the ground when about to spin their cocoons 

 beneath the turf. When the larvae had reached this stage one 

 frequently found that the Rooks had riddled the turf round the 

 bases of the trees with holes in search of the larvae. 



Birds do not normally care for such resinous larvae as those 

 of N. erichsoni, but the present case is an example of a pheno- 

 menon which sometimes occurs when an insect assumes un- 

 natural proportions and becomes a pest, and where the presence 

 of a large r mount of food temporarily alters the feeding habits 

 of many of the birds of that locality. 



During the winter months when the larvae are enclosed in 

 their tough cocoons they are safe from the attacks of their 

 avian enemies, but it was discovered as a result of observation 

 and experiment that the small Field Vole, Microtus agrestis y 

 Flemming (Fig. 1), proves itself of very great service in extract- 

 ing the larvae from the cocoons and eating them. This small 

 rodent burrows beneath the turf and litter round the bases of 

 the trees in search of the cocoons ; having found a cocoon it 

 nips a small piece off one end and draws out the enclosed larva,* 

 which it then devours. The presence and activity of the voles 

 is attested by the numbers of empty cocoons gathered in small 

 groups in the runs that they have made and the teeth-marks 

 of the voles can be readily distinguished on the empty cocoons 

 (see Fig. 2). During last winter (1907-8) these animals 

 destroyed large numbers of the larvae in all the plantations on 

 the Thirlmere watershed which were badly attacked. From 

 observations made in different plantations I calculated that 

 they had destroyed about 25 per cent, of the pupating larvae. 

 Further, dissections of their stomachs and microscopic examina- 



* The saw-fly larva remains in the larval stage in the cocoon until about May of 

 the following year, when it changes into the pupa. 



