184 The American Naturalist. [February 
alcohol by the action of fresh water, or a solution of chromic acid 
in distilled water. A trace of hydrochloric or nitric acid added to the 
alcohol dissolves some crystal deposits, but not those produced by 
calcium sulphate. _ 
The greatest obstacle in the way of obtaining satisfactory prepara- 
tions of Siphonophores is the tendency to split up into fragments 
which many of them, especially those with nectocalyces, show. Fried- 
lander experimented with various salts in an attempt to discover a 
reagent which would kill without producing fragmentation, and ob- 
tained the best results with ammonia, zinc sulphate, and copper sul- 
phate. The first of these reagents is, however, unsatisfactory for other 
reasons. 
An interesting observation in connection with the use of these re- 
agents is, that to obtain good results the reagent must have a certain 
minimal concentration ; below this fragmentation occurs, increasing 
in intensity with the weakness of the reagent. This does not seem to 
depend on the rapid killing or fixation of the tissues by the strongest 
solutions, since such reagents as concentrated corrosive sublimate 
and strong nitric or acetic acids are much more rapid in their action 
than 15 per cent. copper or zinc sulphate, and yet produce excessive 
fragmentation. 
The following method of preservation is recommended for the 
Siphonophores. The vessel which contains the animal in sea-water is 
held in a tilted position, so that the water is almost at the brim on one 
side. A solution composed of 
Pee ed ees vs 1000 Pa 
Oe ene ee a aaa a E 
pee ns a a aa a a 
is then poured in somewhat gradually, so that it may mix equally with 
the sea-water. The amount of the reagent to be used varies with the 
species under treatment., For instance, with Physophora it should be of 
about equal volume with the sea-water ; but for Forskalia, which has 
numerous nectocalyees, it should be at least double that volume. 
After the animal is completely dead, it should be placed in a fixing 
solution, for which Friedlander recommends 1 per cent. chromic acid 
in sea-water, with the addition for more delicate forms of strong Osmic 
acid, or else } per cent. osmic acid alone may be used. — 
Before transferring to alcohol the animal should be allowed to slip (the 
nectocalyces going first) into a glass tube, open only at one end. This 
opening should be plugged with cotton, and the tube suspended, the 
