MOSQUITO BIOLOGY 183 



swamp was frozen solid. January 23, I broke the ice on likely pools 

 and made about a dozen dips in as many places. Nothing was found 

 and the inference is that the insects shelter either in the mud or under 

 the overhanging vegetation. January 30, during a mild spell, collec- 

 tions were made in pools from which the ice had just disappeared, and 

 now quite a little series of specimens was obtained, some of them in- 

 active and apparently dead, but all revived when brought indoors. On 

 the thirty-first, I covered another part of the same territory and found 

 specimens everywhere: "Little holes, not over two inches in diameter, 

 full of water, turned out two or three specimens." February 1, col- 

 lected additional specimens and brought in some of the ice to see 

 whether larvae were frozen in it. I found that there were none and 

 that the specimens probably kept down below actual frost, if possible. 

 February 7, it was noted that there was considerable difference be- 

 tween the larvae, as though two or more stages were represented, and, 

 as a whole, they were more advanced than in December ; in the larger 

 specimens a distinct shield-like form of the thorax was now apparent. 

 Collections were made throughout February, and the indoor cultures 

 were under constant observation ; but though the larvae seemed to feed 

 continually they did not grow and spent most of the time at the bot- 

 tom of the jars ; they do not need air from above the surface. On Feb- 

 ruary 22, when the ground was snow covered and everything frozen 

 over, holes were made in the ice and active larvae were dipped up from 

 beneath it. The water here was 36°, and in the woodland springs 

 where other specimens were taken it ran 42°. Rain and snow so filled 

 the swamp area that the wrigglers were scattered over so large a ter- 

 ritory it was almost impossible to find them. At this time a series of 

 the specimens was sent to me to be developed in the laboratory, and 

 these arrived in good condition. Throughout March collections were 

 made, and during this time it became so dry that it seemed as if all 

 the larvae must have perished ; yet after a rain there were as many as 

 ever. They seem able, apparently, to seek out the wettest places and 

 may even survive for a time in soft mud. April 2 and 3 another lot of 

 over 100 larvae was collected for shipment to me, and up to that time 

 there had been no pupa seen. The first specimen in the culture material 

 pupated April 4, and as I obtained the first pupa just a day earlier, 

 this may be considered the beginning of the pupation period. The pupa 

 is small in proportion to the size of the larva and the period of this 

 stage is from six to ten days. The last larvae were collected May 5, 

 and in nature only one pupa was taken, an accidental one mingled with 

 canadensis. 



