ClCINDELID^: AND CARABHXE 159 



ing, opaculate; legs, antennae and sides of the prothorax dark testa- 

 ceous; head somewhat evidently more than half as wide as the prothorax, 

 the antennal joints long and slender; prothorax relatively rather small, 

 not quite as long as wide, the sides broadly arcuate, converging anteriorly, 

 parallel basally, the basal angles rounded; surface broadly convex, dis- 

 tinctly concave and reflexed at the sides, more broadly and feebly basally; 

 elytra oval, convex, three-fifths longer than wide and fully two-fifths 

 wider than the prothorax, the striae fine, the intervals flat, the micro- 

 reticulations but little longer than wide, irregularly hexagonal, rather 

 convex and shining, the incised separating lines deep; sides evenly and 

 rather : trongly arcuate; pectination of the tarsal claws much shorter than 

 in any of the eastern species. Length (9) 9-5 mm.; width 3.7 mm. 

 California (Los Angeles Co.). 



Allied rather more closely to obscurus Lee., than any other, but 

 more elongate, with relatively much smaller prothorax and more 

 inflated elytra; the antennae are more elongate and the prothorax 

 much less transverse, being in fact nearly as long as wide. There 

 are before me extended series of all the California species except 

 behrensi, which is represented by two examples only. Quadricollis 

 is very abundant in the coast regions from Humboldt to Sta. 

 Barbara; obscurus is rather less abundant from San Francisco to 

 San Diego; ruficollis is by far the stoutest species and occurs abun- 

 dantly about San Francisco; I do not happen to have it from any 

 other locality. My two examples of behrensi are from San Diego 

 and Calaveras Co.; it very much resembles ruficollis, except that 

 it is less stout in build ; guadalupensis is by far the largest American 

 species of the genus; it is confined so far as known to the island of 

 Guadalupe; quadricollis also occurs on the island, but apparently as 

 a recent importation and is not endemic as in the case of guada- 

 lupensis. 



Pristodactyla Dej. 



In general habitus such species as dubia and impunctata, differ 

 very decidedly from any of the above mentioned members of 

 Calathus proper and, on this ground alone, it w r ould seem advisable 

 to retain the genus; but there are certain more specialized structural 

 characters as well, such as the unmargined intercoxal process of the 

 prosternum in Pristodactyla. The Munich catalogue does not 

 recognize Pristodactyla and makes it a synonym of Odontonyx 

 Steph., but the latter is sunk as a synonym of Calathus in the 

 recent European catalogue, which gives to Pristodactyla full generic 



