We quite agree with Pocock's opinion that Palæophonus was not adapted for land and aërial respiration, but "lived in the sea, probably in shallow water, its strong, sharply-pointed legs being fitted for maintaining a secure hold on the bottom."

In conclusion, then, we would suggest, from our present knowledge of the Palæopoda, that the group is a natural one, that the line of descent of the phylum from some Annelid-like worm was independent of that of the crustacean phylum, and that the affinities shown by morphology and embryology to exist between the Trilobita, Merostomata and Arachnida are so close that they form a tolerably definite series of interrelated classes.

Phylum II. PANCARIDA. Represented by a single class, Crustacea. The phylum name is proposed for the reason that the group is so well circumscribed, none but the genuine Crustacea or Carides belonging to it, forms as to whose position in nature all zoologists are well agreed.

In this group there is a decided advance over the Palæopoda in the differentiation of the appendages into from three to six kinds, with corresponding functions. In the lower or more primitive, though somewhat modified, group of Cladocera, such as Daphnia, there are two pairs of antennæ, a pair of mandibles, of maxillæ, and of legs or trunk-appendages, the whole performing four different functions; while in the Decapoda there are besides the antennæ, mandibles and maxillæ, three pairs of maxillipedes, five pairs of thoracic and six of abdominal legs, or appendages, in all performing six different kinds of functions-a degree of differentiation and specialization not exceeded by any other Arthropodan group.

The members of the phylum show an increasing tendency, as we rise towards the more specialized forms, to a heteronomous segmentation and also to a wonderful transfer of parts headwards (cephalization), the cephalothorax being covered by a carapace formed by the hypertrophy or excessive development of the tergites of the second antennal and mandibular segments. In the Phyllocarida the cephalothorax is covered by a bivalvular carapace, with a weak adductor muscle; while in Apus it, in adaptation to its burrowing in soft mud, assumes the general shape of the shield of Limulus; while in the Estheridæ the entire body is protected by the two valves, which are connected by a definite hinge and ligament. On the other hand, the head-shield of the Palæopoda, as well as the pygidium when occurring, is the result of the simple fusion of the

segments. While the Phyllopoda are generally regarded as the most primitive group-their swimming limbs closely resembling those of Annelid worms-it may be questioned whether the Phyllocarida are not a still more primitive group; certainly they are the most composite or synthetic, and were the earliest known group of Crustacea.

Crustacea are, like the Palæopoda, prosogoneate; but when we take into account the fact that there is in the adult but a single pair of nephridia (the green gland), or in other forms (Phyllopoda) the shell gland, there has been a great reduction in the number of pairs from what may have been the ancestral type, while Limulus still retains four pairs. In all Crustacea the eggs are carried attached to the body of the parent, and never, as in trilobites and merostomes, deposited loosely in the sand.

In their metamorphosis, which is a complete one in all the typical forms, the larval stage of the lower Crustacea being a nauplius, we have another feature wanting in the Palæopoda. As is well known, the early embryo of Moina passes through a prenauplian stage like that of Annelida, and the indications are that the nauplius is itself a derivation from the trochosphæra stage of Annelid

worms.

Now, as is well known, the most primitive groups or members of a group do not undergo transformations; and in this respect the Pancarida are a later, more specialized group than the Palæopoda. It will be remembered that the most primitive insects (Synaptera) do not undergo a metamorphosis, and in several of the lower orders of winged insects it is incomplete, there being no larval and pupal stages; in the Arachnida only the extremely modified Acarina undergo a slight metamorphosis. That of the Meropoda is slight.

Enough has been stated, we think, to show that the Palæopoda are quite remote from the Pancarida, and that a union of the trilobites in the same class with the Crustacea brings about an unnatural association, and tends to an unnecessary amount of confusion.

Dr. Kingsley regards the Trilobita as the more primitive subclass of Crustacea, but we are unable to see any features in Crustacea which could have been derived from trilobites; there are no transitional forms, and the larval forms are widely distinct, as he has well shown. The gap between the two groups is, on morphological and embryological grounds, a very wide one. Already in the early Cambrian seas trilobites were a predominant type, while

the Crustacea were comparatively scanty in numbers, and represented by primitive types showing no trace of trilobite characters.

The Chief Factors in the Evolution of Classes.-Assuming that the Arachnida, represented by the most typical form, the scorpion, have evolved from the class of merostomes, in the way suggested by Pocock, the entire process or phenomenon has the most direct and instructive bearings on the method of evolution of one class from another.

In the first place, the single group of scorpions-say a single generic form-appears to have arisen from some genus of Eurypterida, allied to Eurypterus, and by divergent evolution the great class of Arachnida, with its eight orders, appears to have originated by one step after another from a single type, not necessarily an individual, but many, all the members of the genus being modified by similar causes, in the same manner and at the same time.

The modification of a marine Eurypteroid form, perhaps living in a shallow, land-locked basin, perhaps finally becoming brackish, into a terrestrial scorpion, was due to changes in the environment, in the topography; this reacted on the Eurypterid and resulted in change of habits, and consequent adaptation to brackish, and perhaps to fresh, water, and finally to land. With little doubt, all the forms inhabiting the area underwent the same kind of modification and similar adaptation to a new medium, the same changes of function resulting in the disuse of organs adapted for marine existence and the evolution of structures adapting the animal for terrestrial life.

The changes by which the connecting links (Palæophonus) became transformed into a genuine scorpion, the ancestor and founder of the Arachnida, were the following:

1. The loss by disuse of the abdominal swimming appendages (except the pectines), and the ingrowth and reduction by disuse of the expodites, the gills attached to them being carried in, forming eventually the book-lungs of the scorpion, each with its spiracular opening, adapted for aërial respiration.

2. The four hinder pairs of cephalothoracic appendages became slenderer after the animal had left the water and adopted a life on land, under stones or the bark of trees, etc., and the single stout claw of the original Palæophonus became by use, in climbing trees, etc., two-clawed, like those of all Arachnida and insects.

PROC. AMER. PHILOS. SOC. XLII. 173. K. PRINTED JUNE 11, 1903.

3. The compound eyes of the Merostomes became broken up into groups of single eyes.

4. Most remarkable changes took place in the internal organs, resulting in the development of salivary glands, none occurring in Crustacea and other marine Arthropods.

5. The acquisition of Malphigian or urinary tubes which exist in terrestrial Arthropods, Arachnida, insects, etc., but in no marine Arthropods.

6. A gradual reduction in the number of pairs of nephridia, all Arachnida having but a single pair, Limulus having four pairs, and the Eurypterida presumably as many.

7. After the scorpion type became fixed and the spiders arose, the number of pairs of book-lungs became reduced from two pairs in Mygale to one in other spiders, and then began an evolution of tracheæ from dermal glands—a process seen in certain terrestrial planarian worms as well as land-leeches.

8. While the arterial system of Limulus, owing to its localized organs of respiration, is remarkably developed, in the scorpion the arterial system is greatly reduced, and in the tracheate Arachnida, such as the spiders, there are no arteries or venous lacunæ.

It is most probable that the evolution of the Palæophonus descendants, viz., the scorpions of the Carboniferous-assuming that they were true scorpions-took place with comparative rapidity, i. e., by tachygenesis, without the extremely slow method postulated by the natural selectionists, the modification suggested above having contemporaneously affected all the individuals, many thousands or tens of thousands alike. The method was not, as Darwin imagined, the result of a single chance variation gradually and by numberless intermediate forms passing into a species which gave origin to many others, from which were gradually evolved new subgenera, genera, subfamilies and so on, but the method was radically differThe Palæophonus, an Eurypterid, became, we take it, in a comparatively few generations the parent of a scorpion, the representative of a distinct class. The class characters, great as are the differences, especially in its internal organs, between an Arachnid and a Merostome, were assumed with comparative suddenness. New classes, like new species, did not arise from a single but from a large number of individuals. This was Lamarck's doctrine, and it has been reaffirmed by De Vries. This shows that even classes

ent.

are in a degree artificial or ideal conceptions. And so it was with the evolution of mammals from theromorphous reptiles, and of birds such as the Archaeopteryx from reptiles. With our present knowledge we can trace an almost exact parallel between the tachygenic origin, by change in the medium, inducing changes in habits and the functions, of flying in sects from Synapterous forms, that of the Arachnida from the Merostomes, of Amphibia from Ganoid fishes, of reptiles possibly from Amphibia like the Labyrinthodonts, of birds from dinosaurian reptiles, and of mammals from theromorph reptiles (unless the Amphibians, as some contend, were the source of mammalian life).

The exciting causes of the differentiation of classes, as well as orders, families and genera, were geological and topographic changes, enforced migration and consequent isolation, adaptation to a new medium, to new conditions of life, such as a change from marine to fresh water, from fresh water to land, and in the case of pterodactyls, birds and insects, from a terrestrial life to one spent partly in the air.

The early Paleozoic ages as well as the Precambrian were periods of the rapid evolution of phyla, and of class and ordinal types, as shown by Hyatt, the writer, and others. Indeed, it would seem as if the evolution and differentiation of varieties and species succeeded rather than preceded the formation of genera and higher groups. It may be questioned whether the natural selectionists could make any progress in evolution, so to speak, by beginning with merely simple variations, although after the higher or more general groups were originated, and this was by far the most difficult and important step, specific variations set in very rapidly, as early as Cambrian times. Few, except palæontologists, appear to appreciate the rapidity with which evolution in Precambrian and Cambrian times must have operated among the plastic forms which here and there crowded the early paleozoic seas.

Phylum III. MEROPODA. This group is proposed to include the classes of Pauropoda, Diplopoda and Symphyla.

Prosogoneate myriopods, in which the body is in the typical forms cylindrical, the trunk-segments variable in number, but usually numerous, and each segment "double"-i. e., united by a dorsal plate, which was originally two plates which had been fused together (Heathcote), unless we adopt the views of Kenyon that the alternate plates disappeared, the remaining plates overgrow