The Eurypterida of New York/Volume 1/Phylogeny

From Wikisource
Jump to navigation Jump to search


Since the date of Laurie's commentary on the relations of the eurypterids among themselves [1893, p. 520], two important lines of evidence have been obtained bearing on the genetic relationships of the genera. One is afforded by the growth stages of the principal genera, collected at Otisville, the other by Strabops of the Cambric. In the preceding chapter we have analyzed the ontogenetic characters and have fixed upon a small number of these that may, with a fair degree of certainty, be considered as recapitulations from eurypterid ancestry. As these characters are actually expressed in Strabops, it seems safe to use the two sets of facts for a conception of the prototype of the eurypterids, and to proceed from this prototype in tracing genetic relationships. Aside from our knowledge of the one Cambric genus, the geologic succession of the genera is of little aid in the problems before us, since nearly all genera seem to be present in the Lower Siluric.

In the preceding chapter we have recognized as palingenetic characters of the larval eurypterids the approximation of the compound eyes to the margin, the prominence of the ocelli and their tumescences, the lack of differentiation of the abdominal segments and the smaller size of the telson. All of these characters, save the prominence of the ocelli, are typically shown in Strabops thacheri, or the Strabops stage.

While the characteristics of this stage will be the basis of our reconstruction of the prototype, others may be safely inferred from a comparison of Strabops with the general characters of the Siluric eurypterids and from theories of the development of the arthropods in general.

In Strabops the carapace [pl. 1] is short, and a little less wide than the greatest width of the body. It is so in most of the later eurypterids, notably in Eurypterus, Hughmilleria and Pterygotus. The eurypterid carapace has been formed by the consolidation of six segments, and a priori it should not be longer than six of them. As a matter of fact it approximately corresponds, as a rule, to the five first tergites in length, suggesting a condensation. This it does also in Strabops and since the segments in that form are all remarkably short, the carapace is much
The Eurypterida of New York diagram np 1.jpg

Diagram of the phylogeny and geological distribution of the genera of eurypterids

shorter than wide. In all eurypterids the preabdominal segments are greatly wider than long and it is hence safe to infer that the carapace also was originally wider than long, and to assume for the prototype a small, short carapace of the width of the preabdomen.

In one section of the eurypterids the lateral eyes are marginal and faceted, and in the other situated on the dorsal surface and smooth. It is important to determine which of the two forms of eyes is the more primitive. Laurie has argued in favor of the eyes of Pterygotus, representing the first group, basing his view on a comparison with the trilobites (on the assumption that the eurypterids are derived from the trilobites), the Scorpionidae and Thelyphonidae, "which must be derived from some way down the eurypterid stem." The Lower Siluric fauna is constituted of species with submarginal eyes and this fact seems to favor Laurie's view.

It is a proper assumption that the appendages of the prototype were undifferentiated. Those of Strabops are unknown with the exception of two segments of the last pair which indicate a primitive form of appendage, and eurypterid larvae have also failed to afford any conclusive evidence bearing on this point. The Siluric genera all exhibit far-reaching differentiation in the legs; this manifests itself especially in the last pair which has been variously transformed either into broad and strong paddles or excessively lengthened; the preoral appendages have been developed into enormous pincers in at least one genus, Pterygotus, but the four pairs of walking legs which lie between these extremes have remained relatively simple in construction, especially so in Pterygotus where all four are alike, rather slender and without spines. Laurie has regarded the simple character of the Pterygotus legs as an argument in favor of the primitive character of the genus. While we do not share this view we agree as to the manifest primitiveness of these four pairs of legs and therefore assign such appendages to the prototype.[1] In Eurypterus the fourth pair has also retained its primitive character; and while the legs are of uniform length in Pterygotus, in all other eurypterids, mature and adolescent [E. remipes, pl. 2], they form a series that is longer backward.[2] It is therefore to be concluded that, even if the legs were originally of uniform length, a differentiation in length took place very early. The relatively great length of the few segments observed in Strabops indicates that the last pair of legs had attained considerable length and that the differentiation had already commenced in the Cambric. We have accordingly restored Strabops with a series of gradually lengthening slender, nearly spineless legs and would provide the prototype with a similar only more uniformly long series of legs.[3]

The chelicerae, or preoral appendages, are of identical structure in all—only excessively enlarged in Pterygotus—and therefore not available for inferences as to the prototype except that it had them as seen in the majority of the forms.

The opercular appendages have not been observed in either the larvae or in Strabops and are therefore useless for the present inquiry. Laurie has suggested that the median lobe of the genital operculum in Pterygotus shows a less degree of development than in Slimonia and Eurypterus, but qualifies his suggestion by admitting that the details of this structure are hardly sufficiently known to admit of our attaching very much morphological value to it. Hughmilleria exhibits the same type of opercular appendage as Pterygotus but is in all other respects a more primitive form than the latter. The evidence from both Hughmilleria and the appendages of such forms as P. anglicus indicates that the development of the organ in question has taken a different course in the Pterygotus and Eurypterus branches.

We have observed no facts in either ontogeny or in Strabops that would seem to indicate the date of the development of the genital operculum. Laurie [op. cit. p. 525] has argued that both Limulus and the scorpions came off from the eurypterid stem before the great development of the genital operculum, because in Limulus an appendage of the second abdominal segment is present which has become reduced in the eurypterids and in the scorpions the second segment is well developed. The scorpions have, however, not been traced farther back than the Siluric, and Limulus not even so far. Thus they leave us in doubt as to the date of the suppression of the second body segment in the eurypterids, but it is obviously safe to assume that it had already taken place in Strabops in view of the typical development of all other eurypterid characteristics in that early genus.

The abdomen of the prototype of the eurypterids was more or less terete, contracting gradually and thereby lacking the distinct differentiation on dorsal view into preabdominal and postabdominal regions by any abrupt constriction at the posterior end of the preabdomen. This is evidenced by both the adult Strabops and the larvae of the Siluric genera. The segments were of nearly equal length, lacking the marked lengthening of the caudal part of the body which finds its extreme in Eusarcus. The number of the segments was already fixed in the Cambric progenitors, Strabops having six tergites and six postabdominal segments. The larvae of the Siluric genera have less in the nepionic stage but soon reach the full complement, thus attesting in their ontogeny the early fixation of the number of segments in their phylogeny.

The telson exhibits two very distinct lines of development. It is lanceolate in most genera, notably Eurypterus, Dolichopterus, Eusarcus and Hughmilleria, becoming in extreme cases styliform, as in the Stylonuri and some of the earlier Eurypteri and the later Anthraconectes; then again it is broad and bilobed in the subgenus Erettopterus. Laurie leaves the question open as to which style of telson is the more primitive, the geological succession having given him no clue. He suggests, however, that as the pointed telson is characteristic of the earlier trilobites, it also is of the eurypterids. Our investigation supports this suggestion. Strabops possesses a short and blunt pointed telson which fully corresponds with the telson of the Siluric larvae, indicating that this is the primitive form of that organ. The primitive Hughmilleria possesses a similar telson, though it already exhibits a tendency to a broadening and flattening of the proximal portion. Slimonia has this tendency still more developed, so that the telson appears as lanceolate with winglike lateral extensions of the anterior half. A reduction of the posterior spine may then be conceived to produce the telson of Pterygotus, and a further suppression of the axial lanceolate portion would lead to the bilobed telson of Erettopterus. The telson of Hughmilleria, Slimonia, Pterygotus and Erettopterus thus seems to present a continuous series of developmental stages, and the pointed or bluntly lanceolate-triangular form to be that of the prototype of the eurypterids.[4]

Genealogy. In the preceding reconstruction of the prototype we have indicated our views of the genetic relationships of some of the genera, and will now proceed to trace the development of the different branches of eurypterids upward from this prototype as expressed in the appended genealogical tree.

Strabops is such a generalized type that it is eminently fitted to serve as prototype; since it is the earliest eurypterid known it may be an actual progenitor of most Siluric forms. Its relatively great width will be disregarded here as not characteristic of the prototype. It suggests that Strabops is already nearer to the Eurypterus than to the Pterygotus stock.

The Siluric and later genera of the eurypterids distinctly fall into two groups or stocks, namely, that of Pterygotus and that of Eurypterus. The former group (Hughmilleria, Slimonia, Pterygotus, Erettopterus) is principally distinguished by the marginal, faceted eyes; the latter, containing the genera Eurypterus, Dolichopterus, Echinognathus(?), Eusarcus, Drepanopterus and Stylonurus, by the smooth, intramarginal or dorsal eyes. There are many differences in the two which indicate that they separated early and developed as independent stocks. We have, therefore, here separated the Pterygotidae from the Eurypteridae.

First in regard to the Pterygotidae: We have elsewhere analyzed the genetic relationship of Hughmilleria to Pterygotus and Slimonia as shown by the position and character of the lateral eyes, the larger chelicerae, the genital appendage and the telson, and have intimated that in the characters of these organs Hughmilleria clearly evinces a more primitive structure than the other two genera. This view is further supported by the more slender form of the abdomen and the shape of the carapace; the latter while more elongate than that of Pterygotus exhibits a distinct approximation to it in outline. The close relationship of Hughmilleria to Pterygotus is further evinced by the small size of the swimming legs and the form of the metastoma. The endognathites of the two differ, those of Hughmilleria and of the Eurypteridae being spinous, stout and increasing in length backward. The endognathites of Slimonia, except the antenniform first pair, differ similarly from those of Pterygotus, and both Hughmilleria and Slimonia lack the immense development of the chelicerae of Pterygotus, and we believe that the endognathites of these two genera have assumed these characters in order to perform a part of the functions of the chelicerae of Pterygotus and so served as organs of offense and defense, while those of Pterygotus remained undeveloped, being overshadowed by the great chelicerae. As a corollary we may assume that the earlier Hughmilleriae approached Pterygotus in the character of their endognathites.

The genetic relationship between the three genera of the Pterygotidae we conceive as shown in the genealogical tree. An unknown series of forms branched off from the prototype in Cambric time and led in late Lower Siluric time to Hughmilleria. From this was developed on one side Pterygotus proper which again produced the subgenus Erettopterus with bilobed telson, and on the other side Slimonia, which retained the small chelicerae but specialized in developing the second pair of appendages for a tactile function, in the greater development of the opercular or genital appendage, the peculiar exaggerated spinosity of the other endognathites, the lobed and lanceolate telson and other features. Slimonia, represented by but a single species, has all the features of a local and aberrant type.

The Eurypteridae are represented by the far larger stock. Strabops points the course the development of that stock took from the prototype through the Lower Siluric. From the latter era we have only the endognathites on which the genera Echinognathus and Megalograptus are based. These clearly represent another aberrant branch which, judging from the character of the multispinous endognathite, ends here, is independent or may lead to Stylonurus (Ctenopterus).


The Eurypterida of New York figure 25.jpg


The Eurypterida of New York figure 26.jpg
Figures 25, 26 Slimonia acuminata (Salter)
Figure 25 Dorsal view. (From Woodward) Figure 26 Ventral view. (From Laurie)
The main stem of the family Eurypteridae is formed by the genus Eurypterus, whose earliest form seems to appear in the Lower Siluric (Frankfort shale) and it persists into the Permic without materially changing. Corresponding to the vigor shown by its longevity is the great preponderance of its individuals and the number of species it has produced. Among the latter very different stages of progress are found. Some have clearly progressed but little, as shown by their terete, undifferentiated abdomen, short blunt telson and short swimming legs. Such forms are E. maria Clarke, E. pygmaeus Woodward and E. minor Laurie. The form and small size of all these suggest an immature condition or an arrested development. This is emphatically true of the peculiar type for which Laurie [1899, p. 588] has proposed the genus Bembycosoma which has a semicircular carapace and a short, conical body, stout, short telson and apparently but nine abdominal segments. This form, represented by a single species, resembles nothing more than our eurypterid larvae from Otisville and it is very possible that these and some other primitive-looking small species of Eurypterus are cases analogous to those occurring among the recent crustaceans where development has been arrested by parasites which prevent molting and result in dwarfed adults retaining larval characters.

One species of Eurypterus, viz, E. kokomoensis [pl. 25, fig. 1] is of special phylogenetic interest as pointing the way which has led from Eurypterus to Dolichopterus, Drepanopterus and Stylonurus. These genera form a group by themselves, well defined by a number of peculiar characters, the most important of which are the great lengthening of the legs, especially of the last pair; the slender body without distinct differentiation of preabdomen and postabdomen and the slender, styliform telson. Connected with the enlargement of the cephalothoracic appendages is that of the carapace which is relatively longer than in all other eurypterids, shows a distinct tendency to become squarish, and develops a broad border. Likewise the greater development of the legs affects the metastoma which in all these genera is very long and subrectangular with deep frontal emargination. The ninth segment of the fifth pair is not rudimentary as in Eurypterus but fully developed. The epimeral pieces of the postabdominal segments are more prominent than in the other groups of genera.

Eurypterus kokomoensis possesses all these characters in an initial condition. While it still has the swimming legs of an Eurypterus, they are longer and slenderer than in other Eurypteri and the ninth segment has grown out into a spur like that of the feet of Drepanopterus and Stylonurus. All the other endognathites are also longer than in typical species of Eurypterus and the fourth pair is not distinguished from the others by its lack of spines and its slenderness. The carapace already exhibits the characteristic large size, squarish outline and broad border (doublure?) and the telson is distinctly styliform. We have for this reason deemed this species worthy of subgeneric distinction and erected the subgenus Onychopterus for its reception.

From this prototype of the Stylonurus branch two genera, Dolichopterus and Drepanopterus, are clearly derived. Dolichopterus [see restoration, plates 40, 41] is principally distinguished by the development of the ninth segment of the sixth appendage which here forms a broad suboval lobe, instead of the claw of Onychopterus, exactly corresponding to the palette of Eurypterus. The same tendency to the broadening of the spines into lobes or plates is displayed on the fourth endognathite [pl. 45, fig. 2], giving this leg an aspect strikingly different from that of all other forms. All legs in the genotype are powerful organs with large coxae and the cephalothorax is of corresponding size; the gnathobase is especially large and the metastoma long, as in the whole Stylonurus branch. The genital appendage, however, resembles most that of Eurypterus. We may consider Dolichopterus as a more specialized genus derived from Onychopterus.

Like Dolichopterus the genus Drepanopterus is also derived from Onychopterus or from forms most nearly represented by that genus. The fifth pair of legs has become still more lengthened and its segments are round or tubular instead of flattened into a paddle; and the ninth segment forms a strong claw corresponding to the size of the whole leg. The legs form together a series that increases regularly in length backward, and the first four pairs are spinous. The carapace is large, the body slender and the telson distinctly styliform.

Laurie [1899, p. 582] has shown that his genus Drepanopterus differs from Stylonurus only by a negative point of chief generic importance, namely, the fact that its last pair of appendages are not excessively elongated, and he states that Stylonurus developed from Eurypterus by way of forms most nearly represented by Drepanopterus, in which there was greater specialization of the fifth appendage, and reduction of the sixth appendage from the typical digging leg to a purely crawling one. The Otisville material and some of Laurie's drawings indicate that in Stylonurus the second and third pairs of legs are also much elongated and specialized by the multiplication of the spines and by their development into broad lobes in species such as S. excelsior [see restoration plate 47]. The body has become still more slender and the slender form of the telson has given the genus its name.

Stylonurus alone of the whole branch has lived into the Devonic era, attaining there immense proportions. It represents an extreme of specialization that is strongly contrasted to that of the Pterygotus branch.

The main stem represented by Eurypterus has persisted with little change into the Carbonic and even into the Permic. The prevailing expression of Eurypterus in the Carbonic is, however, that represented by the subgenus Anthraconectes. We have elsewhere fully shown that the species referred to this group exhibit distinct phylogerontic characters in the excessive spinosity of the body due to the development of the "scales" into spines, in the elongation of the epimeral pieces and in the excessive length of the telson.

A small independent branch that came off from Eurypterus is represented by Eusarcus. In its triangular carapace, anterior eyes, broadly oval preabdomen and narrow taillike postabdomen, anteriorly longest series of endognathites, curved telson and other features, this genus bears marks of a peculiar aberrant type that apparently left no successors.

  1. Laurie has also advanced as a point in favor of the primitive condition of Pterygotus "the apparently much greater development of the epicoxite—a structure common to the eurypterids, Limulus and Scorpion, and therefore probably primitive—in Pterygotus than in the other genera" [op. cit. p. 521]. The evidence in regard to Eurypterus since obtained by Holm shows that the epicoxite is as strongly developed as in Pterygotus.
  2. Eusarcus makes a partial exception, inasmuch as the second pair is the longest and the following legs are successively shorter; but as the first pair of legs is the shortest, it is obvious that this reversal is a secondary modification.
  3. In order to get a transitional form from the eurypterids to the larval Ammocoetes, Gaskell [Origin of Vertebrates, p. 242] has "made the four endognaths small, mere tentacles in recognition of the character of these appendages in Eurypterus" [see text fig. 16]. While it is admissible to assume the possibility of such reduction of the endognaths for purposes of a hypothetical phylogeny, the fact should not be lost sight of that we have no evidence whatever of any tendency among the eurypterids toward a suppression of these appendages to mere tentacles, but on the contrary it is very clear that in all branches there is a progressive lengthening and development of the legs, with the exception of Pterygotus where 4 pairs remain primitive, but without reduction. The growth in relative size is most notable in the Stylonurus branch, but is observable in the central Eurypterus stock as late as Carbonic time. Gaskell's figure and description must then have reference to a divergent, early branch of the eurypterids, whose existence is quite unknown.
  4. The remarkable observations of Walcott on the trilobites of the Mesonacidae [Smithsonian Miscell. Coll. 1910, v. 53, no. 6] show that the telson of Olenellus is not a true pygidium but originates from a median spine of an earlier segment by the suppression of posterior segments which are still present in the genera Mesonaeis and Paedeumias. The fact that the telson of Olenellus resembbs that of Limulus has suggested to Walcott the view [p. 246] that while this resemblance does not necessarily indicate that Olenellus was the ancestor of Limulus, "its origin does indicate the manner in which the telson of Limulus may have originated."
    The possibility of such origin for the telson of Limulus and the eurypterids is extremely interesting but we have thus far not become aware of facts, in either the ontogeny of the eurypterids or of Limulus and the scorpions, that would suggest such mode of origin of the eurypterid telson spine; indeed the absence of median spines on the dorsal segments of any eurypterids as well as the very early fixation of the number of segments in the Eurypterida, contrast to the variability of these features in the trilobites and militate against the probability of the origin suggested.