86 LABORATOKY METHODS. 



In the first place, isolation, entire or in part, is necessary. This is 

 accomplished by subdividing the tubes which lead from the iron pipes 

 overhead. An inch or two from the faucet there will be slipped in an in- 

 verted T of glass bearing two tubes ; these in turn may be similarly sub- 

 divided by inverted T's, and so on, the number of outlets supplied by the 

 one original faucet and neck being limited only by convenience. Every ter- 

 minus of a tube, whether delivering water or air, is closed by a glass tip, 

 which not only gives exit to a safely diminished stream, but does no harm 

 to the inhabitants of the jar, as the corrupting influence of rubber in con- 

 tact with salt-water might. These tips are bits of glass tubing cut off as 

 required, melted in a spirit-lamp, drawn to a fine point, and perforated by 

 a hole, which allows the escape of only a thread of water or a bubble of 

 air so small as to cause no disturbance. Eacli man makes these glass tips 

 for himself, bending and twisting them to suit his needs. The rubber tub- 

 ing, too, is a great convenience. It is of various sizes, can be cut into any 

 required length, pieced out by stretching over a joint of glass tubing, fit- 

 ted air-tight upon iron pipes, faucets, glass rods, and the like, and bent 

 about in the most handy and time-saving way. 



Though the water comes clear enough, it will not do to allow the air 

 supplied by the windmill to enter at once the water in which the very 

 delicate organisms are being kept alive. It is likely to contain some 

 moisture gathered on the way, and this moisture is liable to have been 

 charged with iron-rust or some other mineral ingredient. The air from 

 the pipe, therefore, is led first through a large Wonlfe bottle, such as 

 chemists are familiar with, where it leaves its deleterious moisture and 

 goes clean to its work. That this precaution is a judicious one, is shown 

 by the fact that the Wonlfe bottles gradually become clouded within by a 

 deposit of iron and dirt. Sometimes ducts of rubber connect two or three 

 jars to one or more of these Wonlfe bottles and to each other, and so 

 there is a constant circulation among a community of little aquaria, econ- 

 omizing apparatus. All these contrivances together, and two hundred 

 and fifty jars and bowls, can be taken care of at once on these tables, 

 though there are only a score or so of supplying faucets. 



Beckoned by their vitality in captivity, marine animals fall into three 

 categories : 



First. — Those that are large and strong enough to allow water to be 

 introduced in a steady stream directly to their jars, and that do not re- 

 quire any more air than the constant current of water brings; these are 

 the crabs, shell-fish, annelids, and common full-grown shore, animals, such 

 as are ordinarily seen in aquaria. 



