70 Psyche [June 



bare white sand, others, modesta consentanea, generosa, hking a 

 surface sparsely covered with pine needles with slight growth of 

 weedy grass. Almost all species are fond of the ruts in the wood- 

 paths seldom rendered dangerous by passing tires or human feet. 

 The "pepo" spiders were omnipresent, although seeking a space 

 perfectly clear for at least a few inches adjacent to the burrow 

 mouth. The various species of wild bees have a habitat of their 

 own, on sloping land where the trees are sparse enough to admit 

 sunlight and where the clay subsoil is confined many feet below 

 the surface. Ant burrows were, of course, abundant and ubiq- 

 uitous. Here and there, too, were many burrows of whose archi- 

 tects we never learned. 



All methods of studying insects underground have their limita- 

 tions, even their failures. On the whole, as satisfactory as any, 

 has been the plaster cast. This was Mr. Brakeley's own inven- 

 tion and a matter of great pride to him. It has already (a score 

 of years ago) been described in scientific literature, but is seldom 

 practised, for although collectors are numerous, observers of nature 

 are still rare. Besides, it sometimes calls for much labor and 

 always for great delicacy of touch. The Cicindela burrow, from 

 six to fourteen inches deep, presents no great task, but that of a 

 Colletes bee, extending sometimes seven feet vertically, is a differ- 

 ent matter. The late John B. Smith, for twenty-eight years a 

 welcome and frequent visitor at Lahaway, dug out one or two, but 

 for the most part he poured the plaster and sat smoking a cigar 

 allowing the hermit to wield the spade until the pit resembled a 

 small cellar. It was easy for me to do the same. In the first 

 place Mr. Brakeley pronounced me an awkward bumpkin almost 

 certain to smash the cast, and secondly he claimed to be fond of 

 the exercise which kept his muscles limber. 



The method in brief: equal volumes of water and plaster of 

 Paris are mixed quickly in a pitcher. If salt water were used the 

 plaster would "set" so quickly that it could not be poured. It 

 has been suggested that by mixing a couple of tablespoonf uls of 

 liquid glue with the water setting is retarded and the cast is 

 stronger. Even at best the mixture must be poured quickly with 

 unshaking hand, since it becomes too hard for use in less than 

 twenty seconds. Three ounces of mixture will fill any Cicindela 

 burrow. About four and one-half are needed for a "pepo" spider. 

 A Colletes requires seven to nine ounces, the pouring of which 



