Mammalian remains have hitherto been found; but in the freshwater formation of the Middle Purbeck, at the top, namely, of the Oolitic series, as many as fourteen small Mammals have been discovered. These constitute the genera Plagiaulax, Spalacotherium, Triconodon, and Galestes. Another gap then follows, no Mammal having hitherto been discovered in any portion of the Cretaceous series (with doubtful exceptions).

Leaving the Mesozoic and entering upon the Kainozoic period, remains of Mammals are never absent from any of the geological formations. From the base of the Eocene Rocks up to the present day remains of Mammals commonly occur, constantly increasing in number and importance, till we arrive at the fauna now in existence upon the globe.

In the following are given the characters of each order of the Mammalia, with the range in time, and, so far as known, the more important fossil forms of each. The number, however, of known fossil Mammals is so great, and in many cases they exhibit so many peculiarities and divergences from existing forms, that nothing more can be attempted here than to give a brief and general sketch of the paleontological history of the class; attention being drawn, where it may seem necessary, to extinct types of special interest.

CHAPTER XXXVI.

ORDERS OF MAMMALIA.

ORDER I. MONOTREMATA.-The first and lowest order of the Mammalia is that of the Monotremata, containing only two genera, both belonging to Australia-namely, the Duck-mole (Ornithorhynchus) and the Porcupine Ant-eater (Echidna).

The order is distinguished by the following characters:The intestine opens into a "cloaca," which receives also the products of the urinary and generative organs, which discharge themselves into a urogenital canal-the condition of parts being very much the same as in Birds. The jaws are either wholly destitute of teeth (Echidna), or are furnished with horny plates which act as teeth. The pectoral arch has some highly bird-like characters, the most important of these being the extension of the coracoid bones to the anterior end of the sternum. The females possess no marsupial pouch, but the pelvis is furnished with the so-called "marsupial bones," be

lieved to be ossifications of the internal tendon of the external oblique muscle of the abdomen. The corpus callosum is very small, and has been asserted to be altogether wanting. There are no external ears. The mammary glands have no nipples, and their ducts open either into a kind of integumentary pouch (Echidna) or simply on a flat surface (Ornithorhynchus). The young are said to be destitute of a placenta, or, in other words, no vascular connection is established between the foetus and the mother. The feet have five toes each, armed with claws, and the males carry perforated spurs on the back of the tarsus (attached to a supplementary tarsal bone).

As regards their geological history, the Monotremes are not known to be represented at all in past time; and this need not excite any surprise, seeing that the order is represented at the present day by no more than two genera, both confined to a single geographical region. Upon theoretical grounds, however, it may be expected that we shall ultimately discover that the antiquity of the order Monotremata is extremely high.

ORDER II. MARSUPIALIA. The order Marsupialia forms with the Monotremata the division of the Non-placental Mammals. With the single exception of the genus Didelphys, which is American, all the Marsupialia belong to the Melanesian province; that is to say, they all belong to Australia, Van Diemen's Land, New Guinea, and some of the neighbouring islands.

The following are the characters which distinguish the order :

The skull is composed of distinct cranial bones united by sutures, and they all possess true teeth; whilst the angle of the lower jaw is almost always inflected. The pectoral arch has the same form as in the higher Mammals, and the coracoid no longer reaches the anterior end of the sternum. All possess the so-called "marsupial bones" (fig. 336), attached to the brim of the pelvis. The corpus callosum is very small, and has been asserted to be absent. The young Marsupials are born in a very imperfect condition, of very small size, and at a stage when their development has proceeded to a very limited degree only. It is believed that there is no placenta or vascular communication between the mother and foetus, parturition taking place before any necessity arises for such an arrangement. As the young are born in such an imperfect state of development, special arrangements are required to secure their existence. When born, they are therefore, in the great majority of cases, transferred by the mother to a peculiar pouch formed by a folding of the integument of the abdomen. This pouch is known as the "marsupium," and gives the name

to the order. Within the marsupium are contained the nipples, which are of great length. Being for some time after their birth extremely feeble, and unable to perform the act of suction, the young within the pouch are nourished involuntarily, the mammary glands being provided with special muscles which force the milk into the mouths of the young. At a later stage the young can suckle by their own exertions, and

they leave the pouch and return to it at will. In a few forms there is no complete marsupium as above described; but the structure of the nipples is the same, and the young are carried about by the mother, adhering to the lengthy teats.

The so-called "marsupial bones" (fig. 336) doubtless serve to support the marsupial pouch and its contained young, but this cannot be their sole function, since they occur in the Monotremes, in which there is no pouch.

They consist of two small bones, which spring from the brim of the pelvis, and which are merely ossifications of the internal tendons of the "external oblique" muscles of the abdomen.

The Marsupialia may be pri marily divided into the vegetableeating and the rapacious or carnivorous forms the former characterised by the rudimentary condition or absence of the canine teeth, the molars mostly having broad grinding crowns; whilst in the latter there are well-developed canines, and the molars mostly have trenchant edges. In the phytophagous section are the living Kangaroos (Macropodidae), the Wombat (Phascolomys), the Kangaroo-rats (Hypsiprymnus), and the Phalangers (Phalangistida). In the carnivorous section are the true Opossums (Didelphida), the Banded Ant-eater (Myrmecobius), the Thylacinus, and the "native devil" or Dasyurus.

As regards their distribution in time, the Marsupialia probably constitute the oldest of the Mammalian orders. Owing, however, to the detached and fragmentary condition of almost all Mammalian remains-consisting in many cases of the ramus of the lower jaw, or of separate teeth-it is not possible

to state this with absolute certainty.

The oldest known

European Mammal is the Microlestes antiquus of the Upper Trias, only a few teeth of which have been as yet detected. The earliest horizon on which Microlestes occurs is in a "bonebed" in the Keuper of Würtemberg; but it has also been detected in the higher "Rhætic" beds. Professor Owen believes that the Hypsiprymnopsis of Mr Boyd Dawkins, from the Rhætic marls of Somersetshire, is also referable to Microlestes. Upon the whole, it is most probable that Microlestes was Marsupial; and it appears to be most nearly related to the little insectivorous Myrmecobius or Banded Ant-eater of New South Wales (fig. 337).

Nearly allied to Microlestes is a small Mammal, a lower jaw of which has been obtained from the Trias of North America, and which has been described under the name of Dromatherium sylvestre. This little animal (fig. 338) appears also to be Marsupial, and to be most nearly related to Myrmecobius.

Each

Fig. 338.-Lower jaw of Dromatherium sylvestre (after Emmons). From rocks supposed to be of Triassic age, in North Carolina.

ramus of the lower jaw contains "ten small molars in a continuous series, one canine, and three conical incisors-the latter being divided by short intervals" (Owen).

The next Mammaliferous horizon above the Trias is the Stonesfield Slate in the Lower Oolites; and there is no doubt that some, if not all, of the Mammalian remains of this belong to small Marsupials. Four genera of small Mammals are known from this horizon-viz., Amphilestes, Amphitherium, Phascolotherium, and Stereognathus. In Amphitherium (fig. 339), the molars are cuspidate, and the animal was doubtless

Fig. 339.-Ramus of the lower jaw of Amphitherium (Thylacotherium)
Prevostii. Stonesfield Slate.

insectivorous. It is believed by Owen to be Marsupial, and to be most nearly related to Myrmecobius. Amphilestes and Phascolotherium (fig. 340) are also believed by the same high authority to have been insectivorous Marsupials, and the latter is supposed to find its nearest living ally in the Opossums of America. Lastly, the Stereognathus of the Stonesfield Slate is in a dubious position. It may have been Marsupial; but. upon the whole, Professor Owen is inclined to believe that it was placental, hoofed, and herbivorous.

With the occurrence of small Marsupials in England within the Oolitic period, it is interesting to notice how the fauna of that time approached in other respects to that now inhabiting Australia. At the present day, Australia is almost wholly tenanted by Marsupials; upon its land-surface flourish Araucaria and Cycadaceous plants, and in its seas swims the PortJackson Shark (Cestracion Philippi); whilst the Molluscan genus Trigonia is nowadays exclusively confined to the Australian coasts. In England at the time of the deposition of the Stonesfield Slate, we must have had a fauna and flora very closely resembling what we now see in Australia. The small Marsupials Amphitherium and Phascolotherium prove that the Mammals were the same in order; cones of Araucarian pines, with tree-ferns and fronds of Cycads, occur throughout the Oolitic series; spine-bearing fishes, like the Port - Jackson Shark, are abundantly represented by genera such as Acrodus and Strophodus; and lastly, the genus Trigonia, now exclusively Australian, is represented in the Stonesfield Slate by species which differ little from those now existing.

Another singular point of resemblance is established by the