MOSQUITO BIOLOGY 129 



sliort hair tufts on dorsal surface and several others at sides, in addi- 

 tion to long lateral hairs. These have four or five hairs on each tuft on 

 anterior two segments, two each on following up to segment six ; 

 seventh with small tufts only. Eighth segment very short, with eight 

 to thirteen scales arranged in single irregular rows on each side. Indi- 

 vidual scale much elongated, with fringing of short hair at sides and 

 apex. Anal siphon dark brown, short and stubby, with double row of 

 spines, from fifteen to twenty-one in each row; small teeth extend up- 

 ward from base of individual spines. Tuft composed of two hairs termi- 

 nates each row. Ninth segment about as wide as long, with saddle same 

 color as siphon. Ventral brush moderate in size; consists of eight or 

 ten tufts of four or five hairs each ; each dorsal tuft has one very long 

 hair. Anal gills short, about same length as ninth segment. 



HABITS OF THE EARLY STAGES 



Unusually long, slender wriggler, easily recognizable by that char- 

 acter as well as contrasting dark head and tail ; is rather deliberate in 

 movements, usually, and has little of the jerky wriggling that fits the 

 proper name to mosquito larvae in general. It has also the habit of 

 forming a loop, bringing its head almost or quite into contact with 

 the anal siphon and then allowing itself to sink slowly to the bottom. In 

 a jar containing a number of the larvae this habit may be observed at 

 almost any time. Some other species of Aedes do this occasionally; but 

 with triseriatus it seems to be a fixed habit. So, the wriggler seems to 

 be able to remain below the surface for quite a long time, sometimes 

 lying quiet, sometimes feeding among the bottom fragments. 



The earliest record for the larva is May 9, 1903, the last for the 

 year on September 3. All these collections were made in holes in trees 

 and stumps, and such cavities when water-filled are, without doubt, the 

 normal breeding places for this species. A few examples were taken at 

 New Brunswick in a little woodland brook broken into disconnected 

 pools during the summer, and a full brood from an iron kettle that had 

 been half buried in sand, was partly filled with chips, and almost hidden 

 in the grass at Lahaway. These are the only records of finding them 

 elsewhere than in tree cavities. 



This larva, in appearance, is nearer like that of Aedes aegypti, the 

 yellow fever mosquito, than any other of those that occur in New Jer- 

 sey. Whether this indicates a closer relationship than the adults seem 

 to show and, whether at a pinch, triseriatus might do the work of 

 aegypti in transmitting the yellow fever, are interesting speculations. 



Development is slow, especially when the larvae are removed from 



