2 NEOTROPICAL PSELAPHIDAE 



in pselaphids, especially the larger species), and the leaf mold of the forest 

 floor. Within my experience I have found log mold to have a higher yield: 

 the mold should be moist but not wet; easily crumbled but not gritty, and 

 brown to black in color rather than red. The moist decaying vegetation of the 

 forest floor, particularly about the base of trees, is good collecting ground. 

 Hollow stumps with several inches of wood bounding the cavity and with a 

 quantity of fine, black mold within the latter are usually infested with micro- 

 coleoptera. Again, rotting stumps, with only portions of the heartwood left 

 and the roots rotted away, can be overthrown, and beneath the stump and 

 upon its upturned base pselaphids may be taken. 



In other words, dense, heavily shaded forests with a deep leaf mold and 

 abundance of prostrate trees are productive, particularly the deciduous trees 

 such as ash, oak, elm, hickory, beech, and maple. Meadows or dry upland 

 oak-hickory forests are not so rich in pselaphids, but may yield uncommon 

 forms. 



Flood plains which are subjected to an annual inundation make poorer 

 collecting ground, although near the top of the flood-line, in the piled debris, 

 some species are to be had, and the uplands of river valleys, especially north- 

 facing slopes, are good hunting. With experience, the pselaphids can be seen 

 and picked up. Two general methods yield the greatest numbers of individuals 

 — sifting and heating. Leaf or log mold can be sifted with great profit by 

 using screens, a practice proved valuable by generations of entomologists. 

 The wire basket with a handle, used for deep-fat frying, having a quarter- 

 inch mesh is neither too bulky nor inconvenient to use. With this or some other 

 suitable screen or sieve, several square feet of white cloth should be carried 

 so that the material sifted can fall upon the cloth and be easily seen. Sifting 

 has one great advantage: the mold can be sifted in situ, the specimens col- 

 lected and another sample tried. In this way unproductive leaves or debris 

 can be discarded, and on the other hand productive material can be worked 

 again and again. The second method for obtaining numbers of individuals is 

 by the application of a gradient of temperature to the contained sample: the 

 Berlese Method. I have used this method with great success. A simple modi- 

 fication of the famous original, which can be constructed easily and at little 

 expense, is as follows: One or more funnels should be made up of stout gal- 

 vanized iron. These can be of any size but I have found funnels measuring 

 ten to fourteen inches in diameter at the top and from ten to twenty inches 

 long very satisfactory. The spout of the funnel should taper to a tubular piece 

 several inches long and half an inch in diameter. Such a funnel is fixed up- 

 right over a finger bowl of 95 per cent ethyl alcohol. A series of circular screens 

 is made up of varying diameters and mesh. I have found three sizes useful, 

 (1) a small screen three inches in diameter with a sixteenth-inch mesh, (2) a 

 medium screen six inches in diameter with a quarter-inch mesh, and (3) a 

 large screen eight inches in diameter with a half-inch mesh. These screens are 

 placed in the funnel at bottom, middle and top thirds respectively. The sample 

 of mold is then inserted on top of the large screen and a heating unit laid 



