44 O 11 N I T H O L O G Y must have had a common ancestor nearer to them than is the ancestor of any Carinate form seems to need no proof. If we add to these the ^Epyornis of Madagascar, the fossil Ratitx of the Siwalik rocks, 1 and the as yet but partially recognized Strutkiolithus of Southern Russia, 2 to say no thing of G astornis, the evidence is stronger still. Scattered as these Birds have been or are throughout the world, it seems justifiable to consider them the survivals of a very ancient type, which has hardly undergone any essential modification since the appearance of Bird-life upon the earth even though one at least of them has become very highly specialized. No doubt the difficulty presented by the biconcave vertebra of the earliest known representative of the Carinate type is a considerable obstacle to the view just taken. But in the American Journal of Science (April 1879), and again in his great work (pp. 180, 181), Prof. MARSH has shewn that in the third cervical vertebra of Ichthyornis " ve catch nature in the act as it were " of modifying one form of vertebra into another, for this single vertebra in Ichthyornis is in vertical section " moderately convex, while transversely it is strongly concave, thus presenting a near approach to the saddle-like articulation "; and he proceeds to point out that this specialized feature occurs at the first bend of the neck, and, greatly facilitating motion in a vertical plane, is " mainly due originally to its predominance." The form of the vertebrae would accordingly seem to be as much correlated with the mobility of the neck as is the form of the sternum with the faculty of flight. If therefore the development of the saddle shape be an indication of development, as well may be the outgrowth of a keel. However, the solution of this perplexing problem, if a solution be ever found, must remain for future palaoontological or embryological dis coverers. The present writer is far from attempting to decide a question so complicated, though he does not hesitate to say, notwithstanding the weight of authority on the other side, that according to present evidence the probability is in favour of the Carinate having been evolved from a more ancient Ratite type. One thing only is certain, and that is the independent arid contempo raneous existence of each of these great divisions at the earliest period when Birds at all like recent forms are known to have lived. The facts that each of these types was provided with teeth, and that the teeth were of a dif ferent pattern, are of comparatively secondary importance. The three It seems therefore quite justifiable to continue, after the Sub- fashion that has been set, to separate the Class Aves into three primary groups : I. Saururx, II. Ratitx, III. Carinatx the earliest members of the two last, as well as possibly all of the first, being provided with teeth. These three primary groups we may call " Subclasses." 3 Thus we shall have : SAURUR/E, Hackel. Arckseopteryx the only known form. RATIT^E, Merrem. a. with teeth ; . with biconcave vertebra as yet unknown; b . with saddle-shaped vertebrae Hesperornis. 1>. without teeth recent and existing forms. 1 For notice of these see the papers by Mr Davies in the Geological Magazine (new series, decade ii. , vol. vii. p. 18), and Mr Lydekker in the Records of the Geological Survey of India (xii. p. 52).
- Bull. Acad, Sc. St Petersburg, xviii. p. 158; Ibis, 1874, p. 4.
3 Prof. Huxley has termed them " Orders " ; but it is more in accordance with the practice of ornithological writers to raise them to a higher rank, and to call the secondary groups " Orders." There is a good deal to be said in behalf of either view ; but, as in most cases of mere terminology, the matter is not worth wasting words over it, so long as we bear in mind that what here is meant by an " Order " of Aves is a very different thing from an " Order " of Reptilia. classes. CARINA1VE, Merrem. a. with teeth ; a , with biconcave vertebra? - Ichthyornis ; b . with saddle-shaped verte bra as yet unknown. b. without teeth recent and existing forms. We have now to consider the recent and existing forms Orders o: of toothless Ratitx. These were shewn beyond doubt by &*#<* Prof. Huxley to form five separate groups, which we shall here dignify by the name of Orders, 4 adding to them a sixth, though little is as yet known of its characteristics. Of this, which contains the great extinct Birds of Mada gascar, he did not take cognizance, as it is here necessary to do. In the absence of any certain means of arranging all of these orders according to their affinities, it will be best to place their names alphabetically, thus : ^EPYORNITHES. Fain. jEpyornithidx. APTERYGES. Fam. Apterygidx (Kiwi, vol. xiv. p. 104). IMMANES. Fam. i. Dinornithidsz ; Fam. ii. Pala- pteryyidx. 5 MEGISTANES. Fam. i. Casuariidx ; Fain. ii. Dromgeidos (EMEU, vol. viii. p. 171). RHE/E. Fam. Rheidx (RHEA, q.v.}. STRUTHIOXES. Fam. Struthionidx (OSTRICH, p. G2 infra). Some systematists think there can be little question of the Struthiones being the most specialized and therefore probably the highest type of these Orders, and the present writer is rather inclined to agree with them. Nevertheless the formation of the bill in the Apteryyes is quite unique in the whole Class, and indicates therefore an extraordinary amount of specialization. Their functionless wings, how ever, point to their being a degraded form, though in this matter they are not much worse than the Megistanes, and are far above the Immanes some of which at least appear to have been absolutely wingless, and were thus the only members of the Class possessing but a single pair of limbs. Turning then to the third Subclass, the Carinatx, their subdivision into Orders is attended with a considerable amount of difficulty ; and still greater difficulty is presented if we make any attempt to arrange these Orders so as in some way or "other to shew their respective relations in other words, their genealogy. In regard to the first of these tasks, a few groups can no doubt be at once separated without fear of going wrong. For instance, the Crypturi or Tinamous, the Impennes or Penguins, the Strides or Owls, the Psittaci or Parrots, and the Passeres, or at least the Oscines, seem to stand as groups each quite by itself, and, since none of them contains any hangers-on about the character of which there can any longer be room to hesitate, there can be little risk in setting them apart. Next comes a category of groups in which differentiation appears not to have been carried so far, and, though there may be as little doubt as to the association in one Order of the greater number of forms commonly assigned to each, yet there are in every case more or fewer outliers that do not well harmonize with the rest. Here we have such groups as those called Pyyopodes, Gavise, Limicofx, G<dlin:e, Columbx, Anseres, Herodiones, Steganopodes, and Accipitres. Finally there are two groups of types presenting character istics so diverse as to defy almost any definition, and, if it were not almost nonsense to say so, agreeing in little more than in the differences. These two groups are those known as Picarix and Alectundes ; but, while the majority Orders o (. a,-in(i 4 See Ann. Wat. History, ser. 4, xx. pp. 499, 500. 5 On the supposition that the opinions of Dr Von Haast (Trans, ard Proc. N. Zeal. Institute, vi. pp. 426, 427) can be substantiated; but they have since been disputed by Prof. Hutton (op. cit., ix. pp. 363- 365), and for the present it is advisable to suspend our judgment.
