among the Aquafica, and in considering the Struthionine birds a distinct group from the other three.

There is, as far as experience yet extends, a very remarkable correspondence between the nature of the upper wing-coverts and of the carotides communes, which adds to the importance of both these characters, which have generally been but little observed. All those birds that have the large upper wing-coverts of the first row on the cubitus so short that they do not reach beyond the middle of the cubital quills, have only one carotis communis, viz., the sinistra. Those birds in which the above-mentioned wing-coverts form several rows and extend beyond the middle of the cubital quills, have, on the contrary, generally two carotides communes, viz., one dextra and one sinistra. The only exceptions to this rule are Cypselus, Trochilus, Merops, one or a few species of Psittacus, Rhea, Phoenicopterus, Podiceps, and Pelecanus, which, although belonging to the latter category in regard to the wing-coverts, yet have only one carotis communis. This is the dextra in Phoenicopterus. We do not, therefore, hesitate to consider these two characters to be among the most important in judging of the affinity of the birds; and they show with certainty that the birds of prey have not their place at the beginning or at the end of the system.

The Strisores, one of the twelve orders in which we have arranged the class of birds, includes several birds that we formerly considered should belong to the Passeres, from their near correspondence in form with the latter. But as they deviate from them in regard to the upper wing-coverts and the claw of the hind toe, and sometimes even in regard to the carotides communes, we are of opinion that they should be regarded as belonging to a different order. They have been separated from the Passeres by Sundevall and by Nitzsch; and the former has arranged them under the order Coccyges, which, according to him, also includes the Zygodactyli and Columbæ. They are, however, distinct from the Zygodactyli in the nature of their feet, and cannot be arranged under this order without depriving it of its most distinguishing character. They seem also to cause confusion if they are arranged within either of the orders Passeres or Zygodactyli; and we have therefore considered it right to arrange them as a distinct order-Strisores, which name was given to them by Cabanis in 1847. However distinct they seem to be, as well from the Columbine section and the birds of prey as from the Zygodactyli, it is very difficult to find any character that sharply and distinctly distinguishes them from these three orders; and we have been compelled to use a character in the scheme that does not belong to all, although the majority of them possess it. They appear to be an intermediate group between Accipitres, Zygodactyli, and Passeres.

The order Longipennes has generally had a very changeable place in the system, sometimes the first among the swimmers, sometimes the last. When the swimmers are, as here, arranged in two groups according to the form of the beak, their place is, as will be seen from the scheme, unquestionable, as we of course must begin with the Pygopodes. The Longipennes approach these very nearly in the genera Puffinus and Halodroma. Puffinus has, together with Colymbus and Podiceps, a long pyramidal erect process at the upper end of the tibia, and the tarsi are compressed like theirs. The genera Phalacrocorax and Mergus form an intermediate link between the Steganopodes and the Lamellirostres.

First Division or Sub-class.

NATATORES, Illiger; Sundevall.

Upper part of the crus skin that covers the body.

(tibia and fibula) not free, but drawn in within the The basis of the hind toe above that of the ante

+ Kongl. Vetensk. Acad. Handl. 1843, pp. 375 and 376.

* Some of the Longipennes are said to form an exception to this.

rior toes, the hind toe sometimes absent. Legs short; and the anterior toes, sometimes even the hind toes, united by web. The upper large wing-coverts of the first row on the lower arm (antibrachium) extend in all beyond the middle of the cubital quills. All, with the exception of Podiceps, have, as far as is known, two carotides communes.

The bill without laminæ.

Group 1. SIMPLICIROSTRES.
Doubly monogamous.*

"Altrices;" that is, carry

food to their young.

Order 1. PYGOPODES, Illiger.

The legs are placed far back; and the hind toe is, when it is present, free. The wings short, hardly extending to the base of the tail. The tail short, or none at all. Heavy, clumsy birds, that dive well, but walk badly. Note. This order contains the typical forms of Natatores.

Order 2. LONGIPENNES, Duméril.

The legs are not so far back; and the hind toe, when there is one, is free. The wings long, extending more or less beyond the base of the tail. They are generally light birds, and lie, when swimming, shallow in the water, and cannot, with a few exceptions, dive, unless they dart from the air into the water, which power a great many of them possess. They generally fly remarkably

well.

Order 3. STEGANOPODES, Illiger.

The hind toe united to the inner anterior toe by a web, and its base but slightly raised above that of the anterior toes. The wings and tail rather large, the former sometimes pointed and sometimes obtuse. Some of these birds are pelagic, fly remarkably well, and are darting divers; some fly badly, but dive and swim well. The position of the hind toe enables some of them at times to sit on the branches of trees and to build their nests there.

Group 2. LAMELLIROSTRES.

The bill with laminæ. Generally singly monogamous. "Præcoces;" that is, do not carry food to their young.

Order 4. LAMELLIROSTRES, Cuvier.

The point of the upper jaw with a so-called nail of the bill; the other part of the bill covered with a soft skin. The hind toe free. The body generally more or less thick and heavy. The power of flight sometimes moderate, sometimes rather inferior. Those that fly best dive badly, or cannot dive at all; the others lie, when swimming, deep in the water, and dive exceedingly well. Some of the former are rather fast walkers, and approach in this respect the next division.

Second Division or Sub-class.

CURSORES, Illiger; Sundevall.

The entire crus and the lower part of the femur free. The base of the hind toe above that of the anterior toes; the hind toe sometimes missing. The anterior toes, when united by a web, are, with very few exceptions, so united only

* Both the old ones sit on the eggs, take care of the young, and carry food to them. The female alone cares for the young.

The majority of the Ardeida make an exception to this; and these live and build very often in trees.

at the base. The large upper wing-coverts of the first row on the lower arm extend beyond the middle of the cubital quills. They have, with the exception of Rhea and Phænicopterus, as far as is known, two carotides communes.

Order 5. GRALLÆ, Linné.

The legs high, and the lower part of the crus without feathers. The wings well adapted for flying. The pectoral bone with a crista. They generally walk and run with ease or very fast, and mostly live in damp places, near swamps or on the banks of water-courses. The majority fly fast and with ease; some fly badly. They live generally in the middle ("mittlere," Faber) monogamy.t Præcoces. A great number of the Ardeida are Altrices.

Order 6. BREVIPENNES, Duméril.

The wings more or less rudimentary, and not adapted to flight. Pectoral bone without crista. A small number of large birds that run fast, and may be considered typical of the whole group. Their structure exhibits a strong tendency towards the mammalian. Some are said to live in the middle monogamy, others in single monogamy, and others again in polygamy. Præcoces.

Order 7. GALLINE, Linné.

The legs of a mediocre height, and the entire crus feathered. The wings adapted to flying, but generally rather short and obtuse, and more or less bent. They run fast; but are easily fatigued by flying, and then hide among rocks, bushes, grass, &c. Some live in polygamy, but the majority live in middle monogamy. Præcoces.

Third Division or Sub-class.

INSESSORES, Vigors; Bonaparte.

The entire crus and the lower part of femur free. The coat of feathers generally extends at least to the tarsal joint.§ The hind toe with its base on a level with that of the anterior toes,[] and very seldom missing.

Order 8. PULLASTRÆ, Sundevall.

The bill not covered by a cere at the base, but generally naked there, and with an inflated skin at the nostrils. The point of the upper jaw rounded, but very seldom bent down in the form of a hook. Three toes directed forward, and not united together. The large upper wing-coverts of the first row on the lower arm extend beyond the middle of the cubital quills. Two carotides communes. The majority fly very fast; some do not fly so well, but these run fast. The majority live in double monogamy, a few in middle or single monogamy (Penelope,) and a few in polygamy (Crax.) The majority are Altrices, the others Præcoces.

Note. This order is evidently an intermediate group between Cursores and Insessores. The Talegallina, Penelopida, and Didunculidæ exhibit some tendency towards the Accipitres.

*The genus Scolopax deviates from this.

+ Both the old ones attend to their young, but do not carry food to them, letting them, under their care, hunt for their own food.

The genus Ortyxelos, Vieill., is an exception to this.

Didunculus (Pleiodus) deviates from this, and has the lower part of the crus naked. The Cathartini form an exception to this.

Order 9. ACCIPITRES, Línné

The bill covered with a cere at the base, convex towards the point; and the point of the upper jaw bent down in the form of a hook. The legs strong, with three anterior toes, which are not united, and are, like the hind toe, armed with strong bent claws. The wings large, with the large upper wing-coverts of the first row on the lower arm extending beyond the middle of the cubital quills. Two carotides communes. They have a strong power of flying, but run badly, and do not jump. Doubly monogamous. Altrices. Their food consists generally of vertebrated animals.

*

Order 10. STRISORES, Cabanis.

The bill without a cere, hard at the base, without any swollen skin at the nostrils, and of a variable form. Three anterior toes, which are generally united at the base, sometimes there united by a web, and seldom free. The hind toe is at times turned forwards. The claw of the hind toe is smaller than the claw on the middle anterior toe, (Sundevall.) The large upper wing-coverts of the first row on the lower arm extend beyond the middle of the cubital quills. Some of them (Caprimulgus, Coracias, Alcedo) have two carotides communes, and some (Cypselus, Trochilus, Merops) have only one. Buceros is unknown as regards its carotides. Some fly remarkably well, others not so well. legs are short in most of them, and not well adapted for walking. Doubly monogamous. Altrices.

The

Note. A polymorphic group, that shows a tendency as well towards the Accipitres and Zygodactyli as towards the Passeres.

Order 11. ZYGODACTYLI, Vieillot.

Two anterior and two hind toes, or sometimes two anterior and one hind toe,t or one hind toe and three anterior ones, the exterior one of which is turned backwards. The claws compressed. The large upper wing-coverts of the first row on the lower arm, except in the Picide and Bucconida, do not extend beyond the middle of the cubital quills. Some have two carotides communes, and others (Picus, Ramphastos, Cacatua) only one. The power of flying not very good. They generally walk badly on the ground; but a great many of them climb well on the trees, and cling skilfully to the branches. Doubly monogamous. Altrices.

Order 12. PASSERES, Linné; Sundevall.

Three anterior toes and one hind toe, and the exterior anterior toe generally at the base united with the middle one. The claw of the hind toe as large as that of the middle anterior toe; and its long flexor muscle separated from the muscle that bends the claw phalanx of the anterior toes, (Sundevall.) The large upper wing-coverts of the first row on the lower arm do not extend beyond the middle of the cubital quills, and we meet with only one row of greater upper wing-coverts. As far as known, only one carotis communis, or truncus caroticus impar, which arises from the left arteria subclavia. Lively and active birds, with a fast and excellent flight, which move easily as well on the ground as on the branches of the trees. They generally jump on the ground, and seldom Some of them have a separate muscular apparatus for singing in the larynx inferior and a more or less exquisite song. Doubly monogamous. Altrices.

run.

Gypogeranus deviates from this.

The thumb or the proper hind toe, which corresponds with the inner hind toe in the others, is in this case missing, except in the Trogonida.

Note This order embraces the typical forms of the group Insessores, and the birds that generally have the highest degree of development.

In the following tables I have tried to use the most important as well as the most positive and evident characters, but have in this, like others, met with much difficulty of finding such of the smaller groups, or families and genera, in the higher orders. A great many of the characters used are taken from Sundevall; and in the Passeres several from Cabanis. Their validity has first been fully tested. In consequence of the above-mentioned difficulty we find that the place in the system of a form in question cannot always be ascertained from similar tables, as a more minute description is often necessary. It must not, therefore, be expected that these tables should give an infallible ground for the determination of the forms belonging to the respective families and sub-families, but only that they should denote some of the most important characters that form the basis for the groups, and give an easy review of these groups. Such a table shows us most plainly what characters are common and what are not.

As a great many of the exotic generic forms are not well known to me, I do not insist that they can be all arranged under the 69 families and 144 subfamilies here charcterized, and that the arrangement of other families or subfamilies is unnecessary; but I believe that a great part of the genera have been considered. I may mention that the difficulties arising in limiting the families Corvida, Parida, and Sylvida among the Passeres have induced me to make these families more comprehensive than they have been.

It seems that the Epimachini and Paradiseini should together form a separate family; but I have not been able to find any distinguishing characters, common to both, that make them distinct from the Corvida. The family Corvida corresponds with "cohors Corviformes," of Sundevall (Sevenska Foglarna) The Troglodytini include forms of both Troglodytina and Timalina, Cabanis, excepting some with emarginated bill. The other Liotrichida, Cabanis, are given to the Sylvida, partly to Lanini and partly to Sylvini. It seems that the family Brachypodide as arranged by Cabanis should at least partly be included in the last-mentioned sub-family (Sylvini,) which, as it also embraces the Sylviada, Cabanis, is very rich, and contains about 500 species or more. I even include the Vireonina, Cabanis, in the Sylvicolini.