14 [January, 



appeared to have been laid. Tlie presence of large numbers of Ilemiptera of 

 the genus AntJiocoris [sj/lvestris and confii&iis), Tvhere\er there was any cover 

 lor them, appeared to account for the failure to tind eggs and later larvae of 

 an autumn brood. 



The larvae live inside tunnels formed by the leaves curlino so that the 

 edges meet ; in these tunnels are usually- three to six larvae. I was for some 

 time in much doubt as to how the leaves became curled. I thouoht it njiaht 

 be due to aphides, which produced a very similar effect, or perhaps to the 

 gall-gnat (Fcrrisia mar(/i/ieiiitorqiie?i.s). There were, however, seldom any 

 aphides with the piliserra larvae, and the gall-gnat r;ir<!ly affected more than 

 the actual margins. Examination of many leaves inhabited by P. piU>^erra 

 of various ages finally convinced me that they became curled bv the 

 presence of thesawfly, probably the active agent being placed by tJie parent fly 

 with the eggs. 



The young larvae eat the parenchyma of the underside of the leaf bet^^■een 

 the various ribs. Sometimes, when larvjie are ^^tv^, this appears to su])ply ;ill 

 their needs, but usually the larvae, as they get large, leave their home in the 

 tunnel during the night and devour neighbouring leaves. 1 never found jiiiv 

 silk used in the tunnels, which resulted apparently from deformation, thou-di 

 I found larvae once or twice hidden between spun-together leaves, where they 

 Iiad taken shelter after excursions for food, but which were abandoned 

 domiciles of other larvae. 



The leaf-rolling result of the attack o'i Fontania piliserra on the leaves uf 

 the osier is clearly of the same nature as the production of galls on species 

 of Salix by not distantly related sawHies. 



The skin-sculpture of the larva consists of points, each with one or 

 two acute ends. These usually point backwards when pressed down flat. 

 They are arranged in transverse rows, sometimes echeloned. With 26 rows 

 across a larger subsegment the acute apices are not well-developed, but are 

 very well shown on a narrower subsegment with 10 to 12 rows. 



PriojjJiorus tener Zadd. 



The second-brood larvae of this species were rather frequent in my cravdt-n 

 in 1920. I found they went freely into rough cork to pupate, and clos"ed the 

 outer opening with a thin glazed diaphragm, behind v.hich a not very thick 

 ordinary cocoon was made. 1 believe practically all the specimens passed the 

 wmter successfully, eniergiug i„ ,Le «priug. 1 succeeded in getting some 

 females to lay eggs. Tliis sunmrer the spring brood of larvae occurred, but not 

 abundantly. I discovered one larva of the autumn brood and saw indications 

 of a lew others, but they were very scarce, especially compared with their 

 abundance the previous year. 



The eggs are laid in the interior of the petioles of the raspberry I saw 

 the operation once or twice, but did not note the precise action of the saws 

 the leaves selected are those that are little more than half developed who^e' 

 })etioies are still tender and soft. They are deposited in the outer or bark-like 

 layer, about hall-way between the surface and the inuer more fibrous and 

 woo.jy central portion, and are placed longitudinall v in a series, each eo-<^ bein- 



