47 



whereas in O. caspius the dorsal head-hairs are all single, and the scales of the 

 eighth segment are many and arranged in an elongate patch, moreover in 0. currici 

 the single scales have a long central spine whereas in O. caspius the central ones 

 are equal and not differentiated. According to the later report (see p. 43) the larva 

 must he regarded as unknown. 



From the numerous hrackish water pools I have had unlimited amounts of 

 larva? for my ohservations. I have never found a single larva with the characters 

 indicated by Howard, Dyar and Knab for O. curriei. 



According to Howard, Dyar and Knab 0. caspius is a brackish water species; 

 flat salt-marshes being essential to its occurrence. 0. curriei on the other hand 

 inhabits temporary pools throughout the arid regions. 



As all Danish mosquitoes with white rings, involving both apex and base of 

 each tarsal joint, only belong to brackish water pools, and mosquitoes of this kind 

 have never been found in the poor remains of an arid region which our country 

 still possesses, I for my own part would be most inclined to refer all these spe- 

 cimens to one single species 0. caspius (= dorsalis). When I suppose it to be most 

 correct provisionally to maintain the species for the Danish fauna, it is because 

 Edwards, without having got the larva, has done the same for England, has looked 

 over my specimens and indicated them as 0. curriei; further because I must con- 

 fess that this variety, more especially with regard to the colour of the thorax, really 

 is more sharply defined than the other varieties. 



Here as — according to a letter from Mr. Edwards — in England the two 

 species occur in the same pools and simultaneously, but 0. curriei is much rarer. 

 In more than twenty-five samples of mine, hatched in my laboratory, in the time 

 from May to July I have never got more than about fifty 0. curriei as compared 

 with thousands of 0. caspius. 



Geographical distribution: Sweden, England, North America. 



3. O. cantans (Meig). 

 Tab. III. 



The three species 0. cantans Meig., 0. annulipes Meig. and 0. vexans Meig. are 

 in their different colour varieties very difficult to distinguish. Many authors, Stager 

 (1838 p. 554), Zetterstedt (1850 p. 3461) a. o. have been inclined to regard (). 

 unuulipes as a variety of O. cantans. The confusion has been augmented because 

 Ficalbi has redescribed 0. annulipes and 0. vexans, and these descriptions have 

 been adopted by Theobald. 



As far as I can see, the old descriptions of Meigen are much more in ac- 

 cordance with the specimens found in our country than those of Ficalbi, whose 

 descriptions perhaps are more in accordance with specimens from more southern 

 countries. The typical (). cantans with its banded legs, dark thorax, covered with 

 reddish and golden-brown curved scales, the brown abdomen with its pale basal 

 bands is an easily recognizable species among the Danish Culicidse. In the old des- 



