98 PAUL S. WELCH. 



to verify the data and to carry on some phases of the work which 

 otherwise would have been impossible. The observations ex- 

 tended over the months of July and August of each of the three 

 years. Larva? of Bellura melanopyga began to appear during the 

 first week of July and could be found until late in August. They 

 feed on the leaves and burrow lengthwise into the petioles of the 

 yellow waterlily, Nymphcea americana (Provancher) Miller and 

 Standley ('12, p. 78). This lily occurs in considerable abundance 

 in the beach pools and the protected bays of Douglas Lake. It 

 also occurs in almost every sphagnum bog in the surrounding 

 territory. This distribution seems to be dependent upon the 

 fact that only at these places are attained the chief conditions 

 which favor their growth, namely, protection from winds and 

 waves, shallow water, and a more or less mucky bottom. N. 

 americana was very abundant in one of the inlets (Bessey Creek) 

 and the larvae of Bellura melanopyga were correspondingly 

 numerous, the infestation in 1912 being 90-95 per cent. The 

 percentage of infestation in the beach pools and in the protected 

 bays was much lower, not exceeding 25 per cent. This was 

 probably due to the fact that Bessey Creek is much better pro- 

 tected from wind and waves, factors which are unfavorable to 

 the very young larvae. The lilies growing in the bogs of the 

 surrounding region showed no signs of infestation. 



FEEDING HABITS. 



The feeding habits of this larva were studied in considerable 

 detail. These activities vary considerably according to the age 

 of the larva. Two fairly well-defined periods can be recognized 

 (i) a very early period, the leaf-feeding period, or the mining 

 period, which lasts approximately throughout the first two stadia, 

 and (2) the later period, the petiole period, which includes the 

 remainder of the larval existence. 



The Leaf -Feeding or Mining Period. The eggs have not been 

 observed but it is evident that they must .of necessity be laid 

 somewhere on the leaf since, as will be shown in a later discussion, 

 the very young larvae are not efficient swimmers and were often 

 found in places where it would have been impossible for them to 

 get had the eggs been laid elsewhere. Furthermore, at the 



