396 PROCEEDINGS OF THE ACADEMY OF 



abdominal segment 3 mill. Described from 3 tf, 6 9 > some rather imma- 

 ture, some tolerably mature. As might have been expected the second joint 

 of the penis is toothed, and is about three and a half times as long as the 

 third joint exclusive of the claws. Belongs to the group " pallidus " of the 

 subgenus Gnrnphus, from the six described species of which group it is sepa- 

 rated at once, as well as the following, by the black costal vein. 



This insect breeds both in the Mississippi River in North Illinois and in the 

 Ohio River in South Illinois, the pupa crawling out on to the banks to assume 

 the imago form. G. fratemus, G. vastus and G. amnicola, mihi, all likewise 

 breed in the Mississippi River, and it is remarkable that in the same spot and 

 on the same day I have seen fratemus, vastus &nA. fluvialis all coming out of 

 the pupa together in considerable numbers. I suspect that most, if not all, 

 Gomphi breed in running, not in stagnant, water. This may explain the fact 

 of my being able to describe no less than six new species of the genus, all 

 obtained within a few miles of the City of Rock Island, which lies on the nar- 

 row point of land formed by the confluence of Rock River with the Mis- 

 sissippi. The habits of fluvialis are very distinct from those of fratemus and 

 vastus. The two latter haunt the land, often occurring in flocks a mile and 

 more from the river ; and from their frequently alighting they are easy to 

 capture. The former makes long excursions to and fro on the surface of the 

 river, scarcely ever approaching the land except for a second, and then never, 

 so far as I have noticed, alighting. Hence it is exceedingly difficult to cap- 

 ture. My specimens were all obtained by taking them just as they came out 

 of the pupa, and allowing them to live as long as they saw fit, which was 

 generally from three to six days. On June 16, 1861, I took a vastus with the 

 Phryganeide macronema zebratum in its mouth. Fluvialis no doubt feeds 

 exclusively on aquatic insects. 



In the autumn of 1860, Dr. Hagen most kindly sent me copies of the magnifi- 

 cent Monographie des Gomphines and Monographie des Calopteri/gines. With 

 the assistance of the former, I was enabled confidently to announce fluvialis 

 as a new and undescribed species in my premium " Essay on Insects injurious 

 to Vegetation in Illinois," p. 341, (printed in the fourth Volume of the Trans- 

 actions of the Illinois State Agricultural Society,) which was placed in the 

 hands of the Society January 3, 1861. I had previously sent a of fluvialis 

 to Dr. Hagen, not having myself met with the J 1 till 1861, and was much 

 gratified by afterwards receiving a letter from him in reply, dated Feb. 16, 

 1861, in which he confirmed my opinion by stating positively that " it forms 

 a new species." I have therefore felt perfectly justified in affixing a name to 

 a species, which I was the first to announce as new, and which is now for the 

 first time described in print. 



Gomphus amnicola, n. s. 9 Differs from the normal 9 of tlie preceding 

 species only in the following particulars : 1st. The ground color is bright 

 deep yellow. 2d. The vertex is yellow behind the vesicle, and there is a lon- 

 gitudinal rectangular ridge on the centre of the vesicle, giving it the appear- 

 ance of being tri- not bi-tubercled. 3d. The angulation of the front is not acute, 

 but the angle is obtuse rather than square and has its apex much rounded off. 

 4th. The basal fascia of the front above is not bi-, but tri-emarginate, viz. 

 one very small and deep central and one large and wide lateral emargination. 

 5tb. The anterior frontal fascia is obsolete. 6th. The labrum is anteriorly 

 margined, but not vittate, with black. 7th. The central lobe of the labium 

 is fuscous at tip and the lateral lobes are tipped with fuscous inside. 8th. 

 The dorsal carina of the thorax is unusually high. 9th. The medial, cunei- 

 form thoracic stripe is much slenderer. 10th. The narrow yellow line sepa- 

 rating the humeral from the antehumeral stripe is only half as wide as in 

 fluvialis, and is interrupted above ; and as a consequence of this and the pre- 

 ceding difference, the antehumeral is very much wider. 11th. The first stripe 

 of the pleura is either interrupted or abbreviated above, and the second is 



[Sept. 



