930 MOSQUITOES OF NOETH AMEEICA 



as inhabit these situations. The proboscis sheath of the imago is strongly 

 chitinized and the month-parts are therefore unfit for piercing. Neither larva} 

 nor adults are closely allied to any of the other genera, but in many points come 

 nearest to the genus Orthopodomyia. It is a curious fact that it is the larvae of 

 the species of Orthopodomyia that are the principal food of the MegarJiinus 

 larvae, perhaps the exclusive food of the more generalized species, the ones inhab- 

 iting hollow trees. We are inclined to think that they are actually related, the 

 Megarhini having been derived from Orthopodomyia-like ancestors, rather than 

 that they resemble them from convergence due to sharing the same habitat. The 

 reason is that there are several other exclusively tree-inhabiting forms that do 

 not possess the characters that the Megarhinus and Orthopodomyia larvae share 

 in common. In fact this derivation implies a lesser break in the evolution, sup- 

 posing some originally tree-inhabiting Orthopodomyia to have acquired the pre- 

 daeeous habit against another species in the same habitat, than that some species 

 not addicted to trees should invade this domain. 



Agassiz and Scudder cite the date 1825 for the tipulid genus Megarhina of 

 Le Pelletiere de Saint Fargeau and Serville (Encycl. Meth., X, pt. 2, 585), 

 which, if correct, would invalidate Megarhinus Robineau-Uesvoidy as used 

 here. The true date, however, is 1828 and in fact a note in the Encyclopedie 

 M6thodique asserts that Desvoidy's name has precedence. This permits us to 

 use the name in the present sense. 



The eggs are laid singly, floating on the surface of the water. E. E. Green 

 in Ceylon has come the nearest of any author to observing oviposition. He 

 says (in the case of M. immisericors Walker) : " The actual operation of egg- 

 laying was not seen, but the female was observed jerking itself up and down 

 in the air just above the water, and it seems probable that the eggs were shed 

 at that time" (Spolia Zeylanica, ii, 160, 1905). A. D. Hopkins discovered a 

 number of eggs on the surface of water in a rain-barrel (in the case of M. 

 septentrionalis Dyar & Knab, Proc. Ent. Soc. Wash., vii, 4, 1905). The eggs 

 are covered with coarse irregular spinose tubercles which entangle air and enable 

 the egg to float. Green describes the eggs (of M. immisericors) as roundedly 

 elliptical, uniformly spinous; Goeldi describes them (M. hcemorrhoidalis Fab.) 

 as much elongate, fusiform, irregularly truncate at one end, densely granular 

 for only two-thirds of the surface toward the pointed pole (Os Mosquitos no 

 Pard, p. 124, pi. N. figs. 114r-117, 1905) . It seems, however, that his description 

 and figures were made from collapsed egg-shells. Dr. Hopkins's specimens 

 before us appear to have been originally regularly rounded, though they are 

 much distorted in the preserved state and rolled up into an elongate shape such 

 as indicated by Goeldi. 



The larvae are entirely predaceous in habit, feeding on the larvae of certain 

 other mosquitoes that with them inhabit water in holes in trees, broken bamboos, 

 at the leaf-bases of bromeliads and in similar situations. M. septentrionalis, 

 with which we are best acquainted, seems to be addicted to Orthopodomyia 

 signifer, though Aedes tnseriatiis inhabits similar holes in the same region and 

 may be also a victim. The larvae will eat any mosquito larvae ofl'ered them and 

 even each other, but they seem to thrive best on the one species, and often fail 

 to be reared in captivity from the lack of the favorite food. Under favorable con- 

 ditions the larvae of some species also occur in water barrels and other wooden 

 receptacles and even in metallic ones. They are never found in any pools or 

 rain puddles on the ground. The prey is seized with great quickness and 

 swallowed alive as rapidly as possible, the struggles of the victim causing an 

 appearance of ferocious activity, like a dog shaking a rat. Green observes 

 that the larvae are dangerous to each other, and where more than one larva exists 



