merging into one another, yet the type of structure, as shown by cleavage, is constant: particular kinds of molecules severally have particular shapes into which they settle themselves as they aggregate. And though in some cases it happens that a substance, simple or compound, has two or even more forms of aggregation, yet the recognized interpretation is, that these different forms are the forms assumed by molecules made different in their structures by allotropic or isomeric changes. So constant is the relation between the nature of any molecules and their mode of crystallizing, that, given two kinds of molecules which are known, from their chemical actions, to be closely allied in their natures, and it is inferred with certainty that their crystals will be closely allied. In brief, it may be unhesitatingly affirmed, as an outcome of physics and chemistry, that throughout all phenomena presented by dead matter the natures of the units necessitate certain traits in the aggregates.
This truth is again exemplified by aggregates of organic matter. In the substance of each species of plant or animal, there is a proclivity toward the structure which that plant or animal presents—a proclivity conclusively proved in cases where the conditions to the maintenance of life are sufficiently simple, and where the tissue has not assumed a structure too finished to permit rearrangement. The perpetually-cited case of the polype, each part of which, when it is cut into several, presently puts on the polype-shape, and gains structures and powers like those of the original whole, illustrates this truth among animals. Among plants it is well exemplified by the Begonias. Here a complete plant grows from a fragment of a leaf stuck into the ground; and, in Begonia phyllomaniaca, complete plants grow even out of scales that fall from the leaves and the stem—a fact showing, like the fact which the polype furnishes, that the units everywhere present have for their type of aggregation the type of the organism they belong to; and reminding us of the universal fact that the units composing every germ, animal or vegetal, have a proclivity toward the parental type of aggregation.
Thus, given the natures of the units, and the nature of the aggregate they form is predetermined. I say the nature, meaning, of course, the essential traits, and not including the incidental. By the characters of the units are necessitated certain limits within which the characters of the aggregate must fall. The circumstances attending aggregation greatly modify the results; but the truth here to be recognized is, that these circumstances, in some cases perhaps preventing aggregation altogether, in other cases impeding it, in other cases facilitating it more or less, can never give, to the aggregate, characters that do not consist with the characters of the units. No favoring conditions will give the laborer power to pile cannon-shot into a vertical wall; no favoring conditions will make it possible for common salt, which crystallizes on the regular system, to crystallize, like sulphate of soda, on
