230 [October, 



water scattered on its surface. It was never allowed to become so dry that 

 the water stood in drops or would not easily soak in. Under this treat- 

 ment partial success was attained. Four adults emerged in the spring 

 of 1921, as follows : April 8th ( 6 ), April 10th or 11th ( c? ), April 13th 

 (morning, $ ), May 8th or 9th ( $ ). These were killed and preserved. 

 The other five failed to emerge, and later in the year the soil was turned 

 out and their earth-coated cocoons were preserved. 



At various times during the spring and summer of 1921 I returned 

 to the patch of iris in the Botanic Gardens and made the following 

 notes. The times of day are given throughout in "summer time." My 

 first visit was on April 28th at about 6.45 p.m., a brilliant evening, but 

 with cold wind. Sevei'al dozen of the adult sawflies were sitting torpidly on 

 the iris leaves, which were still very short. On April 29th, at about 2 p.m. 

 and in blazing hot sunshine, the insects appeared to number a hundred or 

 two. Though less torpid than on the preceding evening they were all 

 sitting still on the flat faces (not the edges) of the leaves, mostly head 

 downwards, some head upwards, but none across the leaf. When dis- 

 turbed they flew, but reluctantly and only a few feet. Some could be 

 picked up in the flngers, and they have markedly the habit of dropping 

 from the leaves " sham dead " when touched. No signs of pairing or 

 egg-laying were observed. On April 28th four specimens were taken, 

 two being chosen because they looked especially small, and these proved 

 to be cJ , while the other two were 5 ; on April 29th six were taken 

 with a random sweep of the net, and all proved to be d • 



On May 1st, 3-4 p.m. and a bright afternoon, the insects were 

 flying short distances rather frequently without being disturbed. 

 Several 5 were inserting their " saws " into the flat sides of the leaves 

 near the thickened middle part. One made several punctures in rapid 

 succession one below the other, f to | inch apart, but so rapidly that I 

 doubt if it actually laid eggs in them. Another remained for some time 

 motionless with the ovipositor inserted into the leaf. This was the only 

 occasion on which the insects were watched piercing the leaves, though 

 adults were seen on the plants in diminishing numbers at intervals up to 

 May 16th, when they were very few. My next visit was not till May 

 29th, when no adults were visible, but groups of young larvae were 

 found for the first time, some of them having already moulted once, as 

 their cast skins testified. No further visit was made for a long interval, 

 till July 5th, when a number of larvae were seen on the iris and the 

 leaves were found to be much eaten, in some cases only a very narrow 

 strip of the thickened middle part being left for about 9 inches down 



