January 3, 1908] 



SCIENCE 



29 



where they arrived September 14. When ex- 

 amined the next morning but two of the larvae 

 and one pupa of the smaller species remained, 

 the others having been devoured by their 

 larger predatory companions. 



The Megarhinus larvse, which had just 

 passed through the last molt (September 10- 

 11, 1906), were placed in beakers containing 

 tap-water, allowing several individuals to each 

 beaker, and kept in the laboratory. They were 

 fed on the larvae of smaller species, chiefly 

 Stegomyia fasciata and Culex sp., until Oc- 

 tober 10, after which time we were unable to 

 secure a further supply of these out of doors. 



When feeding upon the smaller larvae the 

 Megarhinus larvse swallow their prey bodily, 

 but when practising cannibalism, as was ob- 

 served in two eases, the victim is held in the 

 strong mandibles and slowly devoured. So 

 far as observed these larvae make no apparent 

 effort to pursue their prey, but remain quietly 

 near the bottom of the water until a smaller 

 larva approaches, when with a quick move- 

 ment the latter is seized in the powerful 

 mandibles and speedily disappears. In the 

 several instances observed no effort was made 

 to change the hold, whether the prey was first 

 seized near the head, middle or tail. One 

 larva was seen to seize a nearly mature larva 

 of Culex salinarius near the middle, and with- 

 out loosening its hold to swallow it gradually, 

 the head and tail disappearing together. Con- 

 siderable difficulty attends an attempt at swal- 

 lowing their prey head foremost, as one in- 

 stance was noted where the usual three or four 

 minutes were stretched to more thfjn an hoiir 

 before the still struggling victim finally dis- 

 appeared. The Megarhinus larvae are able to 

 remain entirely submerged for hours at a 

 time, even during comparatively warm weath- 

 er, which fact probably accounts for their 

 habit of wintering in the larval stage. 



Four of the larvae died before December 

 10, 1906, and on that date six of the remain- 

 ing twenty were transferred to an earthen jar 

 containing about three inches of mud, above 

 which was two inches of water. This jar was 

 sunk into the ground almost up to the rim in 

 a wire screen house out of doors and sheltered 



from the direct rays of the sun. Here these 

 six larvae remained throughout the winter. 

 The lowest temperature to which they were 

 subjected was +13° F., on the morning of 

 December 24, 1906. This cold snap continued 

 two days, freezing ice three fourths of an inch 

 in thickness on the water in the jar. On the 

 afternoon of December 26 the ice had thawed 

 around the edges sufficiently to allow it to be 

 lifted out of the jar, when one larva was found 

 with its anal breathing tube frozen fast in the 

 ice. The other larvae were lying on the sur- 

 face of the mud in a semi-torpid condition. 

 One of the larv» died as a result of this freeze, 

 and by March 12 two others were dead. On 

 March 25 one of the remaining outside larvae 

 pupated, and another on the thirtieth. A drop 

 in temperature to +30° F. on the morning 

 of April 2 caused the death of the remaining 

 outside larva and one pupa. The other pupa 

 was so weakened by a freeze on April 15, when 

 the mercury fell to + 26° F., that it died 

 two days later. 



On December 15 five of the fourteen larvae 

 kept in the laboratory were transferred to a 

 beaker, in the bottom of which was about two 

 inches of mud, thus duplicating the conditions 

 of the lot kept outside, except as to tempera- 

 ture. No difference in behavior could be 

 noted, except that the larvae were more active, 

 as would naturally be expected from the higher 

 and more uniform temperature of the steam- 

 heated room. 



Of this lot two larvae transformed to the 

 pupa state, in which condition one died, the 

 other emerging as adult (female) on April 4. 

 The last larva died April 18. 



Soon after the mud was placed in the beaker 

 a number of cyclops and other small water 

 animals were observed swimming about. 

 These were still present this spring, although 

 the mosquito larvae were without food for a 

 period of five months, during which time they 

 continued active, but were never seen to make 

 any attempt to feed upon these smaller forms 

 of water life. Nor did they during this time 

 resort to cannibalism; but when several larvae 

 of Culex salinarius were placed in the beaker 

 on March 12 they were devoured with great 



