244 THE AMERICAN NATURALIST.  [Vou. XXXIII. 
be kept for months or years in an ordinary fernery where the 
atmosphere is constantly saturated with moisture. I have in 
my laboratory a large fernery or zerrarium, about 2 x 3 feet 
square and-2 feet high. The bottom consists of a zinc tray, 
8 inches deep and water-tight. The top and sides are of 
glass and the front side runs in a frame with weights, being 
thus capable of being raised and lowered like an ordinary 
window-sash. In the bottom of this there are about 6 inches 
of good garden soil, in which are planted ferns and other wood 
plants. The surface is partly covered with moss, and here and 
there are placed several stones, the size of one’s fist, and a few 
pieces of rotten stump, arranged so as to give shelter to the 
adults. In one corner a crystallizing dish is sunk to the level 
of the soil. This is filled with water and the bottom covered 
with a little fine sand. Some duckweed, or Salvinia, may be 
placed upon the surface, and a few small stones should be put 
in a dish. At the beginning of the season, after arranging 
everything as above, enough water is poured in to drench the 
soil, and the sunken dish is filled. After this the terrarium is 
self-regulating. The water that evaporates is re-precipitated as 
moisture, and the total loss from the little pond in the corner 
is so slight that it needs replenishing not oftener than once in 
six months. If the ¢errarzum is to support many animals, it is 
better to place a few earthworms, myriapods, etc., in it; and if 
the pond is designed for the rearing of larvæ, supplies of Ento- 
mostraca and a little Spirogyra to feed them with should be 
occasionally introduced. I have tried placing tiny bits of meat 
in prominent places, but they merely mould and have to be 
removed. I have kept as many as 20-30 adults and a dozen 
larvee in my ¢errarium during an entire college year, and several 
times, on clearing it out in the fall after the summer vacation, I 
have found alive and in good condition adults which I had been 
unable to find in the spring, when I intend always to remove 
the animals. It seems most probable that these salamanders 
find enough to eat among the worms and insects introduced 
with the earth and plants, as they always appear in perfectly 
normal condition and contrast very forcibly with Diemyotylus, 
which grows thin and often starves to death when placed under 
