separated collections of crypts over the dorsal region, and are hence placed in different families by Günther. Schismaderma of South Africa is a true Bufo without parotoids, and Paludicola another member of the true Bufonidæ differs from Bufo in this respect, as Bombinator does from Alytes. Calamita, a genus of Hylidæ from the Australian region embraces two species, C. cyanea Daud. and C. dolichopsis Cope. The first of these has a large glandular mass on the scapular and even cranial regions (and as such the type of a peculiar family of Günther, his Pelodryadida) while the second has no glandular enlargements whatever; yet Günther (Zoological Record, 1868) states that they are mere varieties of one species.

To go over all the families in detail, would be unnecessary, as published works have already corrected them. Suffice it to say, that of the fifteen into which the non-tree-frogs are divided, but eight appear by their names to represent natural families, though in character not one of them has a good foundation. Of the nine into which the tree-frogs are divided, but two coincide in extent with modern families, and none in characters. One of the latter embraces but one genus, and is called the Hylapleside, though as Peters has shown by autopsy, the Hylaplesia of Boie is a Bufonid, probably a true Bufo. The number of families and subfamilies which might have been constructed on such bases as the above, on genera discovered since the publication of this catalogue, would be considerable, but naturalists have, with one exception* not availed themselves of the privilege.

If we examine the genera of this system the extent of the work of reformation already marked out, becomes more apparent. Thus in his Ranidæ, of thirteen genera, but five belong with Rana, and one of these, Heteroglossa, would be a Polypedatid according to Günther's system. Another, Stenorhynchus natalensis Smith, was subsequently redescribed as the type of a new genus and species (Phrynobatrachus Natalensis) by the director, although generically undistinguishable from another genus of the same author, Dicroglossus. The Cystignathidae contains nothing but members of that family, or rather relations of Cystignathus, but is a mere fragment compared with the cohorts. that really belong to it. The greater part of all the remaining families of tooth-bearing series belong to it, whether tree-frogs or not. The genus Plectromantis Peters, with small digital enlargements on the fingers only, should be placed here, as it is very near to, if at all distinct from the type genus Cystignathus. The great extent of this family is paralleled by other Neotropical forms, as the Formicariida, the Tyrannida, Characinidae, Chromidida,, etc.

* Mivart, Proc. Zool. Soc. London, 1869, 280. The writer in one or two instances did the same.

As to the "Discoglossidæ," fragments of four families are represented by its five genera, one of which immediately follows, viz: the Asterophrydidae. In the "Brachycephalida there are placed three genera, Pseudophryne, a true Bufonid, and scarcely distinguishable from the genus Epidalea Cope, which embraces the old Bufo calamita of Europe; second, Brachycephalus, which is near Phryniscus and belongs to the Phryniscidae; and third, Hemisus, which is nearer by much to Engystoma, but probably forms the type of a distinct suborder?

It is not necessary to examine the genera further, but I turn to the species, where it is proper to express more favorable views. Thus science owes to Dr. Günther a debt of gratitude for the collections, in his British Museum catalogues, of convenient diagnoses of species, with references to many works and authors not accessible to all. His books thus become manuals, and indispensable as the last compilation of an extensive and scattered literature. But it is not only as a compiler that his works are valuable. His usual conscientiousness in attempting the accurate discrimination of species is most praiseworthy, though we cannot help thinking that his estimate of the value of species is sometimes a little interfered with by national and personal predjudices. English, French, and American authors fare the worst at his hands, and we freely admit that in the latter case his criticisms are often deserved, so far as they relate to some of the naturalists of a generation or two ago. These will, however, compare favorably with those who commenced the work in Europe, as Klein, Merrem, Laurenti, Shaw, etc. His countrymen do not, however, escape, and Wagler comes in for the charge of having described, after the much and unjustly criticised work of Spix, "a badly-figured specimen of Ceratophrys ornata "* a new genus, Hemiphractus. Now Hemiphractus is one of the most remarkable and distinct genera of Arciferous Anura of South America, the type of a peculiar family, and Spix's figure represents the type species very well. Prof. Peters first restored the genus.

as

A serious drawback to the merits of the species work, not only of this, but of Dr. Günther's ichthyological works, is his tendency to ignore species. This view approaches those expressed by Prof. Schlegel in the field of ornithology. Thus distinct species are continually united, and even good genera are not unfrequently found involved in the undigested mass. This probably results from the very poor opportunities of studying the Reptilia and Fishes enjoyed by the author, except

*One of the Cystignathida.

+I do not allude to his "doubtful-species" which he puts to one side as too briefly described, often very justly.

in an alcoholic condition; for the European, of all the faunæ, is the most poorly provided with these forms of life. The North American, one of the best provided in these respects, is almost unknown to Dr. Günther, for it is unfortunate for American students that his works in respect to our fauna are of less value than in any other department. It must, however, be added, that some of the older American authors in this field have been quite as bad in another direction, and in respect to furnishing well contrasted specific diagnoses, exceedingly derelict. This has been especially apparent in those who indulge in the execrable practice of publishing preliminary descriptions to "secure priority. This, unless reduced to a system of analytic keys, is nothing but a hindrance to science, and results in warning all students off the ground but the writer, an object which it is safe to presume, he generally has in view.

The system then, presented by the Catalogue of the Batrachia Salientia (Anura), is a phenomenon in the history of our science, and is to my mind one of the least successful of the attempts of skin-zoology to interpret nature. That I intended my remark made on a former occasion,* that it was a "complete and practically useful system," in a euphemistic, if not a pickwickian sense, would, I should think, be sufficiently obvious to any who should compare my system with it. But it is neither euphemistic nor pickwickian for subsequent authors to follow me in abolishing nearly all its leading features and in newly defining all the groups, and then to declare that they adopt Günther's system with a few modifications introduced by me.

In accordance with Günther's system, he was compelled to conclude that the Anura do not display any of the remarkable geographical relations exhibited by other groups of animals, but are rather varied in relation to latitude. This conclusion I have shown to be most erroneous, and that the Anura of all groups, represent the wonderful faunal relations of geographical areas in the strongest light,-in a way not less distinct than any known order of animals or plants.

As opposed to these valuable results, we have the position, that "zoological classification" should " repose on more external and readily ascertainable characters," and that it is "well to turn to such other (characters) as can easily be observed;† all which we suppose will only interfere with the progress of knowledge where sincerely believed and held.

* On Primary divisions of Batrachia Salientia, Nat. Hist. Rev., 1865.

Mivart, Proc. Z. S. London, 1869, 281-2. This author makes some curious objections to the definitions of some of the groups for example, that of the Arcifera, the parietal fontanelle, etc. The objection is, that these characters mark immature stages of other groups! a point which I have considered in an essay on the "Origin of Genera" subsequently published. It will suffice to state here, that this relation indicates for a character a certain fixed grade of systematic value. The lowest (specific) and some highest, are those that appear earliest in embryonic life.

ART. XXXI-On Foraminifera from the Gulf and River St. Lawrence; by G. M. DAWSON.

By way of introduction to these notes, I may state that the reader will find some account of the curious and interesting animals to which the paper relates, with figures of characteristic examples, in vol. iv, new series, of the Canadian Naturalist, p. 413; and that several species found in the Gulf of St. Lawrence have been catalogued by Principal Dawson, in the same Journal, vol. v, page 188 et seq. The following tables give, however, the only approach to a complete view of the species and their distribution hitherto attempted.

Many of the deeper samples were small quantities of mud brought up in sounding, by Capt. Orlebar, R.N., of the Coast Survey, and by him kindly presented to Dr. Dawson.

The specimens from Labrador were obtained from material dredged by the officers of the Geological Survey; those from Prince Edward Island were from a specimen secured by C. Robb, Esq.; and those from the Bank of Newfoundland were obtained from the late Sheriff Dickson, of Kingston.

The somewhat extensive series from Gaspé Bay was obtained during a dredging expedition in the summer of 1869. The mud was sampled when brought up by the dredge, and reserved for examination, the depth being ascertained as carefully as possible. Several very rich and interesting samples are also from the dredgings of Mr. J. F. Whiteaves, F.G.S., in Gaspé and its vicinity. Mr. Whiteaves has also gone over this material with care, and has detected some additional species.

The means were unfortunately not at hand for ascertaining the temperature at the bottom. But, though there is reason to believe that the water at Gaspé Bay is somewhat warmer than the Gulf of St. Lawrence in general, the mud as it came over the boat's side felt icy cold to the hand, showing even here what a great effect the iceberg-laden Arctic current has on the bottom temperature. The number of species tabulated must not in every instance be taken as a criterion of the relative richness of the localities, as much often depends on the amount of material at disposal. This is especially the case when comparing dredgings with soundings.

The general aspect of the Gulf of St. Lawrence Foraminifera is northern, and in many places closely resembles the fauna of the Greenland coast and the Hunde Islands, as given in Parker & Jones' Memoir.* The Gulf, at least so far as its Foraminifera are concerned, evidently belongs to the Arctic province, the

* Philosophical Transactions, 1865.

limits of which skirt the Banks of Newfoundland and pass from thence southward to Cape Breton.

The refrigeration of its waters depends on the Arctic current, which, entering the Straits of Belle Isle, floods the whole bottom of the Gulf with water almost at the temperature of the Arctic seas. To these conditions the series of collections from Gaspé offers somewhat an exception, and is of a slightly more southern character, both as regards the species represented and the development which they attain. This difference depends on purely local causes, which, while slightly changing the character, give opportunities for a very abundant development of Foraminifers, more especially of the arenaceous forms. Gaspé Bay in no part exceeds 50 fathoms in depth; is about 20 miles in extreme length, well land-locked, and disturbed by no other current than that caused by the ebb and flow of the tide. The depth is not so great as to allow of the incursion of the cold and deep layer to any great extent, and the proximity of land and the shelter thus afforded tend still further to modify its temperature.

The bottom, in most of the deeper parts, is composed of fine sand and mud, and this it is which favors the very large development of arenaceous forms.

Past the mouth of Gaspé Bay sweeps the very strong tidal current of the St. Lawrence, and immediately we pass the shelter of Ship Head and come within its influence, the changes in the Foraminifera become strikingly apparent. The bottom consisting for the most part of clean gravel or coarse sand, most of the arenaceous forms disappear at once, and instead of the abundance of Nonioninas and Miliolas previously found, a very large proportion consist of Planorbulina lobatula, which can hold its own, attached to seaweeds and polyzoans. Polystomella Arctica also becomes somewhat prominent, while the Lagenide and Entosolenidæ appear in abundance.

What few sandy forms do occur are depauperated and composed of very coarse particles. The Foraminifera as a whole however are very abundant, and in some samples dredged by Mr. Whiteaves almost equal in quantity those in the deeper Atlantic soundings.

In the estuary of the St. Lawrence itself, Bulimina pyrula becomes a somewhat common form. Among forms which in the Gulf of St. Lawrence may be mentioned as specially characteristic of deep water, are Nodosaria (Glandulina) lævigata, Globigerina bulloides, very small; Bulimina, principally B. squamosa, also small; Uvigerina pygmæa, Cassidulina.

From depths greater than 100 fathoms all the Foraminifera are very small and delicate; and Lagenidæ, Buliminidæ, Globigerina bulloides, together with a few depauperated Nonioninæ, constitute the greater part of the fauna. From these depths

AM. JOUR. SCI.-THIRD SERIES, VOL. I, No. 3.-MARCH, 1871.