vein ; the 2nd vein forking soon after the origin of the 3rd, 

 the submarginal cross-vein placed soon after the fork, but 

 the relative spot at which it joins the 1st vein is variable, owing 

 to the varying length of the 1st vein in different species ; the 

 3rd vein always * with a short basal section ; anterior cross-vein 

 at the upper proximal corner of the discal cell which is five- or 

 six-sided, not often more than twice as long as broad ; the 2nd 

 submarginal cell and the 1st posterior cell subequal, or the latter 

 slightly shorter ; anterior brauch of the 4th vein forked or not ; 

 posterior cross-vein generally beyond the base of the discal cell, 

 sometimes as far distant as the middle of that cell; the 5th, 6tli 

 and 7th veins comparatively straight, the latter sometimes a little 

 sinuate, all turned down somewhat at tip. 



liange. Probably world-wide, though none seem to have been 

 recorded from Africa. 



The difficulty, in fact, the impossibility of breaking up this 

 genus into subdivisions has been fully dilated on by Osten 

 iSacken.t The difference in length of the autennte in the sexes 

 of some species, as compared with their uniform length in both 

 sexes in other species, has no systematic value, as it would 

 separate obviously closely allied species. 



The number of the posterior cells is a character of at most 

 specific importance, open to adventitious variation ; the presence 

 of the additional cross-vein in the 2nd basal cell is accordingly 

 to be regarded as of doubtful generic value.J The venation 

 varies to some extent, often in specimens of the same species, 

 especially in the position of the posterior cross-vein and the 

 relative length of the petioles of the 1st submarginal cell and of 

 the 2nd posterior cells. When the discal cell in isolated indi- 

 viduals is adventitiously open, the anterior branch of the 4th vein 

 bears a double fork, similar to that of DoIicJioj^eza, the posterior 

 branch not being forked at all. 



Life-hisfon/. The larval stages are passed in rotting wood, or 

 in growing plants, according to the species, the metamorphoses 

 of several European species having been observed. Perris describes 

 those of L. cUspar, Mg.,§ as occurring in the dry stems oi Angelica 

 sylvestris, through which the larvte drive longitudinal furrows. 

 The larva is cylindrical, shining hvid grey in colour, with a horny 

 black head; its structure closely resembles that of the typical 

 Tipulid larva3. L. imnctatu, Mg., is noted by Schoffer |i as living 



* In all the species I have seen. 



t Monog. Dipt. N. Amer. iv, \). 197. 



\ Tl)e recent PaljBarctic Catalogue, liowever, admits Idioptera, Macq., 

 Daciylulabis, Os. Sac., Ephelia, Sob., and Pcecilostola, Sch., as good genera. 

 All the new species described herein appear to belong to Limnoiihila in a strict 

 sense. 



§ Ann. Soe. Ent. France, 1849, p. 331, pi. vii, fig. 5. 



11 Verb, zool.-bot. Ges. Wien, 1873, 1878, 1886. 



