74 Osten Sacken: on the atavic index-characters 



its position among the Eriopterina. Having got so far, it was 

 casy for me to discover the analogies between Chionea and Trimicra 

 (in the incrassate male forceps, in the legs and the antennae), and 

 thus, by dint of a correct method, to place a spider-like being alongside 

 of a daddy-long-legs! (Compare Monogr. etc. IV, p. 136 and 168; 

 Studies etc. II, p. 196). It scems doubtful whether, without thosc 

 index-characters, the curious relationship between tvvo so dissimilar 

 forms would ever have been found out. 



The genns Dolichopeza has been alvvays described as having 

 the tibiae unarmed, although, as belonging to the longipalpi, it 

 should have spurs at the tip of the tibiae. I have shown (Studies etc. 

 II, p. 157) that such spurs, although very minute, are really present. 



Empodia and spurs on the tibiae are not the only characters 

 that can be used as index-characters among the Tipulidae. One 

 of my conespondents asked nie why I did not place Rhamphidia 

 somewhere near Dieranomyia, because it is so very like it, and has 

 many characters with it in common. But Rhamphidia, besides its 

 16-jointed antennae {Dieranomyia has them 14-jointed), has another 

 character which, insigniticant as it looks, is very persistent. In both 

 the european and the north-american species the marginal crossvein 

 is wanting. In a brazilian species (Berlin Mus.), which I have de- 

 scribed (Studies etc. II, p. 184), the same crossvein is also wanting. 

 The same is the case with an australian species, described by Skuse. 

 More than that: the genus Elephantomyia is nothing but a Rham- 

 pMdia with a prolonged proboscis and modified palpi; and the 

 marginal crossvein is likewise wanting in Elephantomyia. Next 

 comes Toxorrhina evidently related to Elephantomyia, but distin- 

 guished by the venation in which the second vein appears suppressed, 

 as if connate with the first vein and the costa, in consequence of 

 which the submarginal cell has disappeared. It looks as if the 

 absence of the marginal crossvein in the two other genera had forc- 

 shadowed the contact of the two veins which the crossvein, before its 

 disappearance, had kept asunder. I have attempted an explanation 

 of this anomaly in the Monogr. etc. IV, p. 112; the ligures of the 

 wings will also be found there (Tab. I, hg. 5 Elephantomyia; hg. 6 

 Toxorrhina). 



Thus these three genera form an inseparable group, which I could 

 not place among my hrst section Limnobina, on aecount of the 

 abnormal number of the joints of their antennae and other aberrant 

 characters (as I cxplained above). I placed them therefore in a pro- 

 visional section Limnobina au o mala, which I had formed for a 

 number of genera having a Single submarginal cell, like the true 



