the four radial tubes (n) from common junction at the apex of the cavity to the bell rim. Their length is about equal, and their course in the bell walls is direct, without division or bifurcation. In the interval between the point of union of the covering-scale and nectocalyx, suspended from the under side of the former, hangs a flask-shaped body (p) which resembles very closely the feeding polyps of Agalma. It contains the stomach, and at its free. end is found the mouth. The stomach cavity is in direct communication with the cavity of the covering-scale. From a point near the origin of the polypite there is suspended a long flexible highly contractile tentacle. This tentacle can be wholly retracted at the base of the polypite, but when the Eudoxia is in motion, is found reaching far behind the point of suspension, gracefully extending to a great length. In addition to the polypite we also find a cluster of bells (g) occupying the interval below the covering-scale and its point of attachment to the nectocalyx already mentioned. These bells enclose in their cavities, in place of a proboscis, a globular mass of eggs. It will be seen that the Eudoxia, which I have described, has female sexual bells only; the male bells I have never been fortunate enough to find. The sexual bells are found in all stages of growth, from a simple bud to a well developed bell hanging from a stout peduncle. The history of the growth of the egg after it is dropped from the female bell, will be treated of in a special paper on the embryology of Diphyes. The anatomy of Diphyes seems to me to sustain the homology of the Siphonophores as pointed out in our account of the anatomy of Agalma. The absence of the float at the extremity of the stem offers no difficulty to this homology when we recollect that the air bladder itself is only a modified medusa bell, and consequently homologous to the anterior of the two bells of Diphyes. The posterior nectocalyx is homologous to a true nectocalyx, while the anterior represents the float of the Physophoridæ. The axis of Diphyes, as that of Agalma, is homologous with the proboscis of a Lizzia, and from its sides bud the medusoid individuals. There is this very important difference between 1According to Gegenbaur, Keferstein and Ehlers, male and female sexual organs coexist on the same Diphyes stem. In the Diphyes which I have studied, that is not the case. The male sexual bell of the American form is unknown to me. Leuckart has also long ago (1853-4) shown that the European Diphyidæ are dicecious (Siphonophoren von Nizza, p. 28. Zoologische Untersuchungen, p. 36). the proboscis of Lizzia and the stem of Diphyes, that while the buds from the former separate without absorption of the stomach walls, the Eudoxia appropriates a section of the Diphyes axis to form essential parts of its body. To my mind there is no difficulty in a comparison of the Eudoxia with Lizzia1 and with the Physophorida. Eudoxia is the adult form of which the Diphyes is the "nurse stage," so that we have here a true alternation of generation as in other medusæ. It is natural, therefore, that the likeness between Eudoxia and Agalma should be a distant one, since the latter genus never passes out of the Diphyes form, or the "nurse" from which the Eudoxia buds. On this account I consider the Physophorida as lower, anatomically and embryologically, than the Diphyida. Like those forms of fixed hydroids, which never drop medusashaped buds, and never, therefore, advance out of the fixed "nurse stage," the Physophoridæ never attain as completely developed a form as the Diphyidæ. They never bud off a gonophore as the medusoid bud is sometimes called, but always remain in the embryonic form. As the Diphyes stage is comparable with a strobila or a budding Lizzia, the Eudoxia is the completed generation comparable with the adult Lizzia which drops the eggs, or the sexual form. The following table exhibits the corresponding parts of Lizzia and Eudoxia: Bell. LIZZIA. Manubrium (proboscis). S Tentacle of a bud from the proboscis, the bell of which is aborted. EUDOXIA. Covering scale (a). Modified medusa bud from the probos- Swimming-bell (b). cis, the proboscis and tentacle of which are lost (aborted). Several buds from the proboscis (young Sexual-bells (g). Lizzie). 1 The comparison of Eudoxia to a “ budding Lizzia" was set first forth, substantially as given in this article, by Professor McCrady in 1857 (Gymophthalmata of Charleston Harbor, p. 67). Since that date the theory has been urged on embryological grounds, without a mention of McCrady's suggestion, by Haeckel, Metschnikoff and P. E. Müller. (Haeckel. Zur Entwick. der Siphonophoren, 1869. Metschnikoff, Stud. uber Entwick. d. Medusen u. s.w., Zeit. f. Wiss. Zool., Bd. XXIV. P. E. Müller, Iag. over Nogle Siphonophorer, Nat. Tidsskrift 3. R. 7. B. Resumé in French). I am indebted to my friend, the late Mr. G. Winther, for a written MS. translation of portions of Müller's work. If we were to follow precedent in our studies of the Siphonophores, we must apply to the adult the name Eudoxia instead of the almost universally used Diphyes. It is just as absurd to retain the name Diphyes to designate anything but a younger stage in the growth of Eudoxia, as it would be to designate the adult sea-urchin a pluteus, or to retain the word auricularia for the adult starfish. The monogastric form, or the Eudoxia, is the adult; the polygastric, or Diphyes, the larva. There is another point to be considered. If from the embryonic feature of possessing a long axis, or stem, the relatives of Diphyes are referred to the Siphonophores, is that reference a good one, and would the characters as assigned to the group to which Agalma belongs (Siphonophora) hold in descriptions of the adult Diphyid? The Eudoxia has no stem-like structure, which gave the name to the group, although it is a true relative. The comparison of Eudoxia with Agalma, or the adult stage of Diphyes with the corresponding larval condition, Agalma, is evidently legitimate, as the comparison of the developed bud of Lizzia with a genus similar to the Lizzia from which it budded. Although we have in Eudoxia an alternation of generation, it is unlike that condition in some other animals, as in the echinoderms, where the nurse cannot be homologized with the adult. In some respects it resembles most closely that process of growth which we are familiar with under the name of strobilation. The Eudoxia is the separated Ephyra, and the Diphyes is a free-swimming scyphistoma, as far as the relation of the nurse to the adult goes. Here however the parallelism ends. The same holds true also in a comparison of genera of Diphyide with the free-living proglottids of the tape worms (Leuckart, Siphonophoren von Nizza, p. 29). As McCrady has already pointed out (Lectures), there is a close resemblance between a Tania and the Scyphistoma in mode of strobilation, but as there is no homology between the proglottids and the Ephyra, so there is little in common in the structure of the Diphyizoid and Ephyra. They resemble each other simply in the mode of strobilation. The corresponding parts of an Agalma and an Eudoxia are given in the table below: The axis or stem of Agalma is reduced in Eudoxia to the polypite condition, and is not distinguishable from this structure. REMARKS ON THE CRETACEOUS AND TERTIARY FLORA OF THE WESTERN TERRITORIES. THE BY LEO LESQUEREUX. 'HE following notes were suggested by two valuable commucations to Nature, in the numbers bearing date June 30 and October 6, 1881; the first, that of Dr. J. S. Newberry, tending to show that the flora of the Dakota group, together with that of the Laramie group, are of Cretaceous age; the second, that of J. Starkie Gardner, Esq., of London, contending to the contrary, that both those floras are Tertiary. As there is not any fixed characters admitted as standard points of determination of the age of a fossil flora, phytopalæontologists have no means of coming to an understanding on the subject, except by a comparison of the vegetable remains of the divers formations with those of localities whose geological horizon has been ascertained. I take here, for comparison with the plants of the Dakota group, the Upper Cretaceous flora of Groenland, Atane; that of Moletin, of Quedlinburg, of the Quader-sandstone of the Hartz and other localities of Germany where this formation, generally considered as Middle Cretaceous, or Cenomanian, has been observed. One hundred and seventy specific forms of plants are now known from the Dakota group; they represent six ferns, one Equisetum, or seven cryptogamous acrogens; seven Cycadeæ, ten conifers, three monocotyledonous plants; the others, about one hundred and fifty, all dicotyledonous angiosperms. As far as known until now, the flora of Atane, Groenland, is represented in sixty-three species-thirteen ferns, two Cycadeæ, ten conifers, three monocotyledonous, while thirty-four, or a little more than one-half, are angiosperms. The relation of the Atane flora with that of the Dakota group is marked by ten identical species: one fern, two conifers and seven dicotyledonous; while quite as distinct an affinity is demonstrated by allied types of the genera Ficus, Sassafras, Diospyros and Sapindus. The flora of Quedlinburg is composed of twenty species; four ferns, four conifers, one monocotyledonous, with eleven angiosperms, a little more than half of the species. Of this group of plants, the relation to the flora of the Dakota group is shown by only one identical species, a fern, which is also found at Atane and Moletin, while analogy is marked by two species of Myrica and a Proteoides. Moletin, in eighteen species described of its flora, has one fern, four conifers, one monocotyledonous and twelve angiospermous plants, these, therefore, constituting two-thirds of the flora. Though the number is small, the flora is related to that of the Dakota group by identity of one fern, one conifer, both also recognized at Atane, and of two dicotyledonous species. This is a remarkably close relationship indeed, more intimate than that between the Quedlinburg and Moletin floras, and it is positive, for the species indicating it, Gleichenia kurriana, Pinus quenstedti, Aralia formosa and Magnolia speciosa, all described by Heer, are of easily identifiable characters. The quader-sandstone of the Hartz is, by its numerous species of Credneria, related to the no less numerous representatives of the genus Protophyllum of the Dakota group. In the Monde des Plantes, by Saporta, the author, who has had opportunity to compare specimens of plants of the Cenomanian of Bohemia with those of the more common and characteristic species of the Dakota group, remarks, p. 202, that the flora of this group presents, if not identical species with those of Bohemia and Moravia, at least a number of equivalent forms. Mr. Feistmantel says, in a note to Professor Heer, that the lower division of the Cretaceous of Bohemia (Perutzer-Schichten) is Cenomanian. After naming a number of plants found in the sandstone of this formation, he adds that the beds of shale, partly between, partly above the sandstone, contain remains of plants, ferns, conifers and a mass of dicotyledonous leaves and freshwater shells. Of the forty-nine species determined by him, nine are also at Moletin, seven at Niedershoena, while three ferns and conifers are present in the Lower Cretaceous of Groenland, and four in the Upper, that of Atane. Of the same plants the Dakota group has five, positively identified: Gleichenia kurriana, Pinus quenstedti, Sequoia Reichenbachi, Magnolia speciosa and Aralia formosa. A sixth might be added, Sequoia fastigiata, but its identification is less definite. And still with the flora of Niedershoena, that of the American Cretaceous is related by one identi1 Fl. Arc., Vol. III, p. 3. |