with design"-we have only to remember the fate of the tapeworm's eggs-that by the larval state the period of infancy and weakness is prolonged, and the period of maturity and efficient care for the continuance of the species delayed, it follows that curtailments and reductions, consequent on adaptation have, as advantageous modifications, a prospect of perpetuation. As in Amphibians the prolongation of the larval phase may be effected by natural circumstances and artificial experiments, so in like manner a compression of the phases of transformation, and a general curtailment of the metamorphosis is imaginable. In the class of Amphibians we have, in fact, several examples of curtailed and modified metamorphosis which bridge over the apparent chasm between development with and without transformation, and render direct development comprehensible as being gradually acquired. Amphibians will endeavour to extend themselves wherever they are invited by a sufficient supply of insects, and the black salamander of the mountains (Salamandra atra) has even overcome the impediment which might have been deemed insurmountable, the absence of water for its larvæ. It does not lay its eggs like its congeners, but only two are received into the oviduct, and the fluids secreted from its walls replace the marsh to them and to the larvæ which emerge from them. Here, and not when separated from the parent, do the gills make their appearance, while the other eggs, gradually following, are devoured by the hungry larvæ. The metamorphosis of the black salamander, which has lately been the subject of Fr. Marie Chauvin's remarkable experiments, can be effected after exclusion, and observation places it beyond a

doubt that this is an instance of adaptation to unusual conditions. If the mode of life of the marsupial frog, which carries its young in a membranous fold of the back, and the Surinam toad, of which the larvæ live singly in the chambers of a kind of honeycomb on the back, were better known than they are, we should assuredly arrive at the same results as with the black salamander. In the absence of other knowledge, the observations of M. Bavey, Marine Pharmaceutist at Guadaloupe, first published in 1873, are of the highest importance. A frog of those parts (Hylodon Martinicensis) goes through its whole metamorphosis in the egg. In the egg it has gills and tail; and from the brief remark that the island contains only rapid running streams, and nowhere stagnant waters or marshes, it appears that this is also a case in which adaptation modifies and curtails development.

66

If, after this introduction, we now examine the socalled direct development with more attention, it may in every way be compared to the metamorphosis of the Hylodes of Guadaloupe. Direct development is a transformation in the ovum; and in the cases in which it occurs, the phases of embryonic development are repetitions, more or less distinct, of the historic development of the family. We will only particularize in the embryonic life of the Vertebrata (in which metamorphosis does not take place), some phases that are stages of curtailed transformation, and recapitulate the permanent condition of their progenitors. It has been repeatedly mentioned that in all vertebrate animals, the vertebral column is first laid out as an unsegmented cord and an unsegmented sheath for the

spinal cord. This is the permanent state of the lower fishes. In the higher Vertebrata also, the brain at first consists of vesicles, lying one behind the other, which is the persistent form of the lower groups. The embryonic heart of mammals and birds begins in the form of a tube, and subsequently acquires the communications between the chambers, which in the reptiles never close. In the Amphibians, the branchial arches really bear gills during the larval state. They are not wanting in the embryos of reptiles, birds, and mammals, any more than the fissures through which, in fish and the larvae of Amphibians, the water passes off after being inhaled. Must we again set forth the only possible explanation of these facts?

Before referring to the phenomena which testify the emanation of families from a common root, we will cite one of the most important evidences of recent times, which traces the genesis of species through a great geological period, and exhibits in detail the relations of the development of the individuals to that of the species, genus, and family. We mean L. Würtenberger's contribution to the geological evidence of the Darwinian theory, to which we have already appealed (p. 97). It relates to the two families of Ammonites, the Planulata and Armata; of which, according to Würtenberger's researches, the latter are developed from the former, as the ribs of the Planulata gradually pass into the spines of the Armata. Of special interest to us are the following passages of the preliminary communication on the discoveries obtained from thousands of specimens, and which will probably not be made public, with all the vouchers, for some years to come. "It gave me parti.

cular pleasure," says Würtenberger, "when, after divers careful comparative studies, I at last detected an interesting and simple conformity to law in the variations of the Ammonites. Namely, on the first appearance of a modification which subsequently attains essential importance in an entire group, it is only slightly indicated on a portion of the last convolution. Towards more recent deposits, this modification is more and more plainly shown, and then advances, following the spiral course of

FIG. 19. Ammonites Humphresianus. A form analogous to the Planulata.

the shell; that is to say, it gradually takes possession of the central turns also, as we trace the forms to higher strata. This reproduction in younger stages of life of modifications first occurring at a more advanced age, makes but slow progress, so that we see the older forms repeated with great persistency in the central turns. Frequently a modification of this sort has taken possession of only a small part of the convolutions, when a new one already appears at the outside, and follows the

first. Thus searching through the strata from below upwards, we see modification after modification beginning at the outer part of the Ammonites, and advancing towards the centre of the discs. The innermost convolutions often resist these innovations with great persistency, so that we usually find upon their surface several of these states of development in close juxta-position, as the shell of the individual Ammonite begins with the old morphological type, and then adopts the modifications in the same order in which they follow in vast periods in the geological development of the groups concerned."

"The Ammonites," he says moreover, "thus obtain at an advanced and maturer age-only when they have gone through the development inherited from their parents, and as much as possible in the same manner as their parents-the power of modifying themselves in a new direction, that is to say, of adapting themselves to new conditions; yet these modifications may then be transmitted to the offspring, so as to appear in cach subsequent generation a trifle earlier, until this phase of development in its turn characterizes the greater portion of the period of growth. But this last and longest phase of development scarcely ever suffers itself to be supplanted by new ones, formed in like manner; heredity operates so powerfully, that a period of development thus once predominant, is repeated in the infancy of the Ammonites, even though but slightly indicated. Hence in an individual Ammonite from a recent stratum, the periods of development compressed and forced back upon the innermost convolutions, must appear in the same succession in which they wrested the dominion