closest embryological study can reveal their composite nature. On the other hand, corals as well as many protists, such as the Labyrinthuleæ, coexist with so small a degree of integration that the parts are considered as distinct individuals, although clearly dependent on one another.
The vegetable kingdom illustrates still more clearly the manner in which the aggregates are compounded. We have plants, like Caulerpa, which, while the form would lead us to expect a considerable degree of organization, consist in reality of a simple aggregation of homogeneous cells. In higher plants, the leaf forms a new order of organization, and constitutes the morphological individual or unit. In trees, the process of compounding has gone so far that, considered as individuals, they may reach the hundredth degree.
If we contemplate the mineral kingdom, we are again shown the same truth. The various recognized minerals are not generally found to be composed directly of the simple chemical compounds into which they may be resolved, but consist of compounds of different orders into which the simpler compounds enter as units of composition. Thus feldspar contains silica, alumina, peroxide of iron, lime, soda, potash, magnesia, water, etc., as units of composition, none of which is supposed to exist in the mineral in any simpler state, and all of which are already more or less complex chemical compounds. Moreover, two or more of these minerals thus formed often again combine as new units to form others of still higher organization.
When we consider the facts which chemistry furnishes, we see the same law still operating in great simplicity. In many of the binary, ternary, and higher compounds, theory requires us to assume that the substances entering into them do so in their integral state, and are not first decomposed into their primary elements and then reorganized into the new compound. The hydrated oxide of potassium, for example, is not written , but , in which both the immediate constituents are regarded as maintaining their composite state and entering bodily into the new compound. The entire series of "compound radicles" requires the same supposition and illustrates the same general principle. Cyanogen (), ammonium (), methyl (), ethyl (), and the rest are now held to constitute integral units in the formation of the hydrides, alcohols, and acids.
So far, then, as induction can be depended upon, we find that it is a universal law of the aggregation of matter that each new aggregate may become a unit for the formation of aggregates of higher orders. Does this law cease with the so-called chemical elements, or are these themselves the products of molecular aggregation?
Without discussing the old and apparently insolvable problem of the divisibility of matter, it may be remarked that while the known facts of science are entirely satisfied with the hypothesis of an ultimate, finite unit of matter, of which all perceptible objects are but