BRACHYGLUTINI 127 



The phylogenetic position of this genus is uncertain and many students 

 have had something to say regarding the species. Raffray (1896, p. 262) said 

 "The lack of suflfieient materials renders the characteristics of the genus erron- 

 eous up to the present", and now, some forty-five years later, the position is still 

 unsatisfactory. In 1850 LeConte erected the genus, based on Eupsenius glaber, 

 the genotype (PI. XIV) ; later this distinguished biologist described another 

 species rufus (1863). Since then seven species have been described from the 

 neotropics. Eupsenius appears to be very isolated in the northern part of its 

 range, has no near relatives in the morphological sense, and both its zoogeo- 

 graphic affinities and morphological affinities lie in the neotropics. 



If Brachyglutini are limited by distant metathoracic coxae in part, and 

 Euplectini in part by contiguous to approximate coxae, then Eupsenius belongs 

 in the latter tribe. If, on the other hand, the Brachyglutini are chiefly limited 

 by triangular coxae and Euplectini by conical coxae, then Eupsenius belongs 

 in the former tribe. If Brachyglutini are said to have globular coxae, then it 

 becomes of interest to determine whether the triangular coxa is a globe or a 

 cone. Final emphasis on coxal shape is of value where one has worked in a group, 

 but is confusing and hardly practical otherwise since (a) exact coxal shape 

 varies between species, (b) a cone is more or less triangular while triangular 

 coxae are more or less conical if they have any length and thickness ; moreover 

 this descriptive term applies to the produced mesial end of each coxa and not 

 to the morphological whole which extends laterally nearly to the margin of the 

 metathorax. 



The separation of the coxae has not been understood generally. Brendel and 

 Wickham (1890, Vol. I, p. 1) describe the coxae of Eupsenius as widely sepa- 

 rated. This is not correct, as pointed out by Raffray (1896, p. 262) , and Raffray 

 (1904, 1908) keyed out the genus as having the posterior coxae subcontiguous, 

 but placed it at the end of the tribe with Barada. However in the recent treat- 

 ment b}^ Bowman (1934, p. 2) Eupsenius will not key out since it is assigned 

 to the section of the tribal key having the coxae "very widely separated". In 

 his generic key, however, Eupsenius will key out perfectly (Bowman, 1934, 

 p. 76) on the Raffrayian characters affecting the ventral surface of the head. 



In later years Raffray (1904, 1908) apparently had decided that this genus 

 was definitely Brachyglutine, but in his last discussion of the genus (1909, p. 36) 

 felt that Eupsenius and Eupsenina were very aberrant types of brachyglutine 

 stock which established a transition between the Brachyglutini and the Tychini. 

 It is readily seen that his earlier remarks (1896) are still available for adoption 

 now. It is even possible that this genus should form a new tribe or subtribe. 



Regarding the antennae of glaber, LeConte originally stated that the 

 eleventh antennal segment was "wider in the middle than the tenth and one 

 half longer". He must have meant something else since the eleventh antennal 

 segment is much larger than the tenth segment, being longer than the third 

 to the tenth segments combined in glaber. 



The two species of the United States are predominantly subtropical in 

 their distribution, having been recorded from South Carolina, Florida, and 



