15 



the apex it bears the anus, surrounded by four anal or tracheal gills, four appen- 

 dages varying in size individually. In ponds situated only a few meters from each other 

 I have found in one of the ponds larvae of O. communis with the gills only half 

 as long as the anal segment; in another larvae with gills more than three times as long; 

 they seem to be most strongly developed in water which is extremely dark and peaty. 

 In some species, f. i. F. geniculate: and C. pipiens, two of them are much smaller 

 than the two others. In F. geniculata they are remarkably broad, in 0. caspius and 

 0. detritus extremely short. Some authors regard these organs as having no respira- 

 tory value, as being only of locomotorical significance, but most suppose that their 

 main function is really respiratory. (Babak 1912 p. <S1 ; Lima 1914 p. 18). 

 Koch (1918 p. 105) supposes that the chief function of the branchial leaflets is 

 probably the absorption of oxygen, while the elimination of carbonic acid from 

 the blood takes place through the body walls. Lima maintains that larvae whose 

 branchial leaflets show numerous tracheal ramifications, remain normally longer 

 under water than those with only small ones, and that the former can live longer 

 than the latter when they have no access to atmospherical air. 



The anal segment further bears the most important organ of locomotion, the 

 large swimming brushes; these are divided into two parts, the dorsal and the ven- 

 tral brush; the dorsal is always the smallest in the Danish larvae; it is inserted on 

 two chitinous pieces; generally it consists of a coarse multiple tuft with rather long 

 hairs; below this tuft are inserted two, often very long, stiff hairs to which, as 

 in T. annulata, a few shorter ones may be added ; in F. geniculata, C. pipiens and C. 

 nigriiulus it is replaced by a few long, stiff hairs. The ventral brush is more highly 

 developed and inserted upon a system of chitin staffs, well described by Howard, 

 Dyar and Knab. "The tufts of the ventral brush hinge upon slender, transverse strips 

 of chitin and towards the middle, where the tufts are inserted, these strips are 

 thickened and perforated. Outwardly these transverse strips are jointed to a pair of 

 longitudinal strips. The tufts are not inserted exactly upon the median line, but 

 alternately a little to one side or the other thus showing the bilateral origin of the 

 structure" (1912 p. 89). The number of the tufts varies from one species to another; 

 it is only about ten in F. geniculata, C. pipiens and nigritulus but more than twenty 

 in T. annulata. Each tuft consists of an undivided arched basal part, bearing on 

 its apex the hairs; this number is fairly constant for the different species; it is only 

 two in F. geniculata, but about twenty in T. annulata; from seven to nine hairs 

 would probably be the most common number. Before the real ventral swimming 

 brush a number of much shorter tufts are often inserted; these are not supported 

 by transversal chitin bands; among the Danish species these tufts are only wanting 

 in Finlaya geniculata and C. pipiens; they are but slightly developed in Ta-niorhyn- 

 chus. Laterally between the dorsal and ventral brush on the above-named shield a 

 short tuft consisting only of one or a few hairs is inserted; among the Danish 

 species it is best developed in F. geniculata. 



Mouth-parts: The labium is divided into three parts: an unpaired median 



