What has made this spider abandon its trap-door nest and adopt the projecting tube habit? The karoo is dry and dusty and wind swept, and so one can easily understand why an underground, doorless hole should have a projecting funnel; but this will hardly account for the abandoning of the trap-door.
One may note, however, that one of the most common spiders here is a Lycosa (L. subvittata) which sinks a hole in the ground with projecting tube with irregular rim. A trained eye can generally detect the difference between these two nests at a glance, though sometimes even such an eye will be at fault; but, to the untrained eye, the projecting tubes are so much alike that they are, in the majority of cases, indistinguishable. Now, this Lycosa belongs to a wholly different family of spiders, in the two-lunged group; it is a less handsome, smaller spider, pugnacious when handled and remarkably active and wary—unlike Hermachastes, which is slow and dull. Can it be a case of imitation, and that Hermachastes has adopted the habit of building a nest like L. subvittata in a part of the country where this tube-building Lycosa is common all over the veld? The interest deepens, as will be seen, when I come to describe the habits of another Lycosa.
There is another trap-door spider, Hermacha (also a new species), closely allied to Hermachastes, which also has a doorless hole but no projecting tube. Its hole, which is sometimes ten inches deep and beautifully lined with white silk, just ends straight off, level with the surface of the ground. Frequently, however, the opening has a delicate, smoke-like web curtain spun across it, which effectually prevents dust getting into it and bars the way to such enemies as do not dig the spider out. Dr. Purcell thinks this habit may be merely the spider's way of shutting itself in when moulting, but it seems to me to occur too frequently and the inmate to be too lively for the acceptance of such an explanation. Hermacha builds in stiff clayey (brak) soil, which cakes like a stone when dry. When dug up, it shows fight, rearing itself up, raising its legs, and throwing forward and parting its fangs so that a bright red gap is exposed between them. The nest of Hermacha was unknown until I found it here.
Passing by several new and interesting species of this family, we come to perhaps its most representative members, the large spiders that make the largest and strongest doors. Of these I have found three species here, two new Stasimopus and one new Gorgyrella (a new genus, recently named by Dr. Purcell.) These two forms are closely allied and superficially bear a strong resemblance to each other, except that, generally, Stasimopus has a darker cephalothorax and legs. They are slow in their movements, large, with powerful digging teeth and stout, strong, shortish legs. Their silk-lined holes are prac-
