ALPHEUS HYATT, the eminent zoologist, died at Cambridge, Mass., January 15, 1902, in the sixty-fourth year of his age.

Outside of his many valuable publications in pure zoology, Professor Hyatt's chief reputation will rest mainly on his researches in the field of organic evolution. No other American has contributed so much toward the discovery of the laws of develop ment and growth, and to an exposition of exact methods of research in evolutionary problems. The principles he enunciated constitute the foundation of a young and vigorous school of evolution, which is already making itself felt in the scientific world.

Alpheus Hyatt was born at Washington, D. C., April 15, 1838. He completed the Freshman year at Yale with the late O. C. Marsh, in the class of 1860; then, after traveling a year in Europe, he entered the Lawrence Scientific School at Harvard, and was graduated in 1862. He served during the civil war, attaining the rank of Captain. Later, he renewed his studies with Louis Agassiz, and has since been intimately identified with all the scientific interests centering about Boston. His official connections were with the Essex Institute, the Peabody Academy of Science, the laboratory of natural history at Annisquam, the Massachusetts Institute of Technology, Boston University, Teachers' School of Science, the Museum of Comparative Zoology, the United States Geological Survey, and the Boston Society of Natural History, of which he was the curator since 1881. In 1869, he was elected a fellow of the American Academy of Arts and Sciences, and in 1875 a member of the National Academy of Sciences. His name also appears in the roll of fellows and members of many other societies at home and abroad.

Among his principal publications should be mentioned: Observations on Polyzoa (1866); Fossil Cephalopods of the Museum of Comparative Zoology (1872); Revision of North American Porifera (1874-77); Genesis of Tertiary Species of Planorbis at Steinheim (1880); Genera of Fossil Cephalopoda (1883); Larval Theory of the Origin of Cellular Tissue (1884); Genesis of the Arietid (1889); Phylogeny of an Acquired Characteristic (1895); Guides for Science Teaching; and numerous essays on the stages of growth and decline in animals, and on various laws and problems of evolution.

C. E. B.

THOMAS MEEHAN, the well known botanist and nurseryman, died on November 19 in his seventy-sixth year. An Englishman by birth, he came to Philadelphia in 1848, and was soon successfully established in the nursery business. He published many horticultural and botanical papers and also the two volumes entitled "Handbook of Ornamental Trees" and "The Native Flowers and Ferns of the United States" (1878-79).

THE

AMERICAN JOURNAL OF SCIENCE

[FOURTH SERIES.]

ART. XIV.-The Ventral Integument of Trilobites; by C. E. BEECHER. (With Plates II-V.)

IN previous papers by the writer, on the structure and appendages of Triarthrus, no attempt has been made to describe or illustrate the character of the ventral integument, especially in the sternal or axial region. The specimens hitherto described were prepared to show details of the appendages, and though portions of the ventral membrane were exposed in many individuals, the subject was not considered of sufficient moment to warrant a distinct study, particularly as no characters were observed in the cuticle that had not been previously seen in more or less perfection by Walcott" in the genera Ceraurus and Calymmene. A recent discovery by Jaekel, however, necessitates the separate consideration of this structure. This necessity arises from the fact that a positive addition to the knowledge of the trilobite anatomy may be deduced, although, as will be shown, Jaekel was apparently entirely misled in his interpretation of the nature of his discovery.

In the paper under discussion, Jaekel states that the occa sion for his publication arose from the finding of a specimen of Ptychoparia striata, from the Cambrian of Bohemia, in which some structures were preserved in the axis of the six anterior segments of the thorax. These, he asserts, are the proximal joints of the legs.

The specimen was preserved as a cast in a rather coarsegrained sandstone, and is exposed from the dorsal side. From certain surface indications of lines in the cast, Jaekel was led to follow these into the rock filling the axis, and succeeded in AM. JOUR. SCI.-FOURTH SERIES, VOL. XIII, No. 75.-MARCH, 1902.

finding a central groove, with two oblique grooves on each side. These he considered as representing the cavities left by the removal of the test from the basal joints of the legs, which thus must have been attached along the median line of the sternum. The supposed joints of the legs were filled with rock, and his attempts to separate them from the matrix resulted in failure.

In the oral region, there were still more indefinite and obscure evidences of cavities left by the removal of some ventral testaceous structure.

These meager remains in the rachis of the thoracic and oral regions have furnished data for what must be considered as the most remarkable and erroneous reconstruction of the trilobite appendages and anatomy that has appeared since the time of Burmeister, in 1843. The latter, in the absence of any material, confessedly based his opinions of the ventral anatomy wholly upon theoretical considerations. Not only has Jaekel to a large degree set aside the evidence presented by many scores of specimens of Triarthrus, as described by the writer, in which each detail of structure can be verified indefinitely, but has also overlooked that afforded by the material illustrated by Walcott, Billings, Mickleborough, and Woodward.12 Moreover, this single specimen of Ptychoparia has led its describer to reconsider on a false premise the entire question of the anatomy, ontogeny, phylogeny, and affinities of the trilobite.

10

6

It is the purpose of the present article to show that numerous individuals of Triarthrus, as well as some material representing other genera, preserve evidence of what seems to be the same structures as those described by Jaekel in Ptychoparia, and also present indisputable testimony as to their correct nature. It will be demonstrated that they do not belong in any way to the appendicular system of the trilobites, but are really the buttresses and apodemes of the ventral body integument. The marvelous state of preservation of many of the specimens of Triarthrus, whose appendages have been studied by the writer, affords very satisfactory indications, not only of the presence of a ventral integument, but also of some of its detailed characters. Jaekel states that in his opinion the unfavorable ("ungünstigen ") preservation of Triarthrus has obscured the proximal structure of the legs, so that what he calls the three basal joints are equivalent to the single unjointed gnathobase of the coxopodite, as described by the writer. Inasmuch as Jaekel has never seen the original specimens described, his statement is practically without foundation. It may also be added that the types and best-preserved individuals have been

retained in the collections of the Yale University Museum. The photographic illustrations accompanying this article, it is believed, will refute his statement, and the specimens themselves would serve the same purpose more completely, since from the black nature of the rock and the nonactinic character of the fossils the photographs feebly represent the delicate structures actually preserved, which are clearly visible to the

eye.

The ventral membrane of Triarthrus, as well as of other trilobites where it has been observed, is of extreme tenuity and only under the most favorable conditions has it been preserved. The membrane itself was a thin, uncalcified, chitinous, flexible pellicle, and thus was in strong contrast with the much thicker and calcified dorsal test.

In the preparation of a specimen to show the appendages from the ventral side, very little of the ventral membrane is commonly exposed, owing to the crowded arrangement of the legs, but when the appendages are removed it is possible to view the entire ventral integument. This process has been carried out in a considerable number of specimens, and some of the more evident characters are herewith described.

The membrane under each pleuron (pleurotergite, Jaekel), or the pleurosternite, as it may be termed,* was smooth and extremely thin, and in the fossils it is invariably concave. This was probably the condition during life, to allow space for the biramous legs and for their infolding during enrollment. It should be noted, however, that the dorsal and ventral integuments in the fossils are generally very close together throughout, leaving but a small cavity for the soft parts of the animal. The space inside has doubtless been considerably reduced by partial collapse from the decay of the soft parts of the animal and also by the pressure of the sediments. The size of the body cavity is unquestionably more correctly shown in the specimens described by Walcott" and Mickleborough, from the Trenton limestone and Cincinnati shales, respectively, where they have apparently suffered less compression.

9

Walcott showed that the membrane in Calymmene and Ceraurus was strengthened in each segment by a transverse arch, to which the appendages were attached at the sides of the axis. These arches were connected by a thinner membrane

* Jaekel has suggested the name mesotergite to supplant the terms axis or tergum, and pleurotergite in place of pleuron or epimerum, as applied to the trilobites. This seems a useful terminology since the older terms are often loosely used and have somewhat different meanings in other groups. Applying this system of nomenclature to the ventral integument, the writer would propose the terms mesosternite for the membrane beneath each mesotergite, and pleurosternite for the membrane beneath each pleurotergite. The interarticular membranes are not included.

(the interarticular membrane), and were aptly compared to the arches in the ventral integument of many of the decapods. Similar features are present in Triarthrus, as illustrated in Plate IV, figure 1, and Plate V, figures 2-4, where it is seen that the interarticular membrane (Plate V, fig. 4) in a normally extended individual is somewhat less than half the length of the arches. The chitinous integument of the arches, or mesosternites, as they may conveniently be called, is thickened along the borders, and appears to be slightly incurved on the posterior edge. The arches are further strengthened by a series of median and oblique longitudinal ridges, or buttresses, which are generally progressively more developed in passing anteriorly from the pygidium along the thorax to the neck segment of the cephalon.

The ventral arch of each segment has the following arrangement of these ridges: There is first a median ridge generally extending from the posterior border entirely across the plate, but sometimes becoming obsolescent near the anterior border. Then, on each side, there is an oblique ridge making an angle of about sixty degrees with the posterior edge and extending inward, but not meeting, the median ridge, thus enclosing a subtriangular space with the anterior apex truncated. Outside of these ridges but still within the axial region there is often a second pair of somewhat more oblique ridges, enclosing rhombic

areas.

The ridges are clearly produced by a thickening of the ventral integument, and can be seen when viewed from the dorsal side of a specimen in which the dorsal test and filling of the body cavity have been removed. They are thus partly or wholly of the nature of apodemes, or plates of chitin, which pass inward from the mesosternites and divide as well as support internal organs, and they are not, therefore, in any sense the proximal joints of legs. Besides serving in this manner they were doubtless efficient in giving the necessary firmness to the ventral arches for the attachment of muscles.

Were these observations confined wholly to the specimens of Triarthrus, there might still be some chance of error, although it is believed that the evidence presented by this genus alone is quite sufficient. Additional data, however, will now be given, regarding other genera and families of trilobites, described independently by other authors, and with no intention of representing the detailed characters of the ventral arches. the search for trilobite appendages by various investigators, the ventral membrane has naturally been of secondary consideration, and in the case of Jaekel's work was of no consideration whatever.

In