Here’s a nice and interesting discussion on how species was seen well before Darwin, which is effectively a “biological”, that is to say, reproductive isolation species conception. It’s by James Prichard, an anthropologist, and published in the second edition of
Prichard, James Cowles (1845), The natural history of man; Comprising inquiries into the modifying influence of physical and moral agencies on the different tribes of the human family (2nd edn.; London: Hippolyte Bailliere). First edition, 1843. Pages 10–18
NATURAL HISTORY OF MAN.
OF GENERA, SPECIES, AND VARIETIES.
THE ancients applied the term genus, or γένος, to any collective number of organised beings which are akin to each other, or the offspring of the same ancestors. The idea of genus was then simple and definite, and just what we attach to the terms kind or kindred. By degrees the meaning of genus was extended, and it was made to comprehend all such creatures as by reason of some real or fancied resemblance in their form or nature were conjectured to have belonged to one original stock. Such groupes [sic] were the dog‑kind, the cat‑kind, the ox‑kind. When it was discovered, in the progress of scientific investigation, that these classes were too comprehensive, and included tribes so remote from each other that they could not with probability be regarded as the progeny of the same original tribes, the term εἰδος, or species, was adopted, and made to express nearly what that of genus had originally denoted. Species was then synonymous with stock or race. But modern naturalists have, as we shall perceive, modified the meaning of species nearly as their less scientific predecessors extended that of genus. They have attempted to found an opinion chiefly on general resemblance, what organised creatures may or may not belong to the same tribe or kindred; and where this seemed admissible, they have termed the aggregate a species. “We unite,” says M. Dc Candolle, “under the designation of a species all these individuals who mutually bear to each  other so close a resemblance as to allow of our supposing that they may have proceeded originally from a single being or a single pair.” “This fundamental idea is evidently built upon hypothesis.”* “The degree of resemblance,” he continues, “which authorises our bringing together individuals under this designation varies very much in different families; and it happens, not unfrequently, that two individuals belonging originally to the same species differ more among themselves in appearance than do others of different species. Thus the spaniel and the danish dog are, as to their exterior, more different from each other than the dog and the wolf, and the varieties of our fruit‑trees offer greater apparent differences than many distinct species of plants.”
Buffon had long ago defined species in similar terms, as “a constant succession of individuals similar to and capable of reproducing each other.” He here combines two circumstances, viz. those of possible reproduction and of mutual resemblance. He had, however, previously observed that the point of resemblance is only an accessory idea; the single circumstance of propagation or of production from the same stock, or in other words that of supposed kindred, or consanguinity, is, in fact, the essential characteristic of species, as it originally was of genus apart from all conjectural extension of the primary meaning of that term. Cuvier adopted nearly the same definition as Buffon. He refers to mutual resemblance between individuals as a criterion of species ; but species itself is fundamentally, according to both these writers, “la succession des individus qui se reproduisent et se perpétuent.”†
* M. De Candolle, “Physiologie Végétale,” tom. ii. p. 689.
† Buffon, “Hist. Nat.” Cuvier, “ Règne Animal.”
It has been acutely observed by a writer who has of late directed his attention to inquiries connected with this  subject,* that the celebrated naturalists above cited have comprehended too much in the definition of species, and, besides laying down what the term species really means, have involved an hypothetical criterion of specific identity and diversity, or of the method of ascertaining the extent and limits of these departments in organised nature.
This remark is undoubtedly well founded, not in regard only to the writers whose names have been cited, but to almost all naturalists. The adoption of a term partly of hypothetical meaning has obviously been the fruitful source of many long and intricate discussions. As the word species, apart from all hypothesis, means only what we express by kind, kindred, τὸ συγγενές, we might avoid a great deal of unnecessary trouble by declining the use of so disputed a term; but as we cannot banish from our vocabulary an expression so well established, we must be content to use it in its proper and restricted meaning as above pointed out.
Species, then, are simply tribes of plants or of animals which are certainly known, or may be inferred on satisfactory grounds, to have descended from the same stocks, or from parentages precisely similar, and in no way distinguished from each other. The meaning of the term species ought always, for the reasons now explained, to have been restricted to this precise import; and when the expression is used in the following pages, it is so to be understood.
The principal object of the following work may then be described as an attempt to point out the most important diversities by which mankind, or the genus of man, is distinguished and separated into different races, and to determine whether these races constitute separate species or are merely varieties of one species.
* M. Flourens, “Annales des Sciences Naturelles.”
Of Varieties, and Permanent Varieties.
Before we enter into the matter of this inquiry, it is necessary to have a clear notion of all the terms that may be used. The meaning attached to the expression permanent varieties approaches very near to that of species, and it is requisite to be careful in distinguishing the two things. Permanent varieties, it being allowed that such tribes exist, are races now displaying characteristic peculiarities which are constantly and permanently transmitted. They differ from species in this circumstance, that the peculiarities in question are not coeval with the tribe, but sprang up in it since the commencement of its existence, and constitute a deviation from its original character.
Some naturalists suppose that many of the tribes now considered as distinct species, both in the animal and vegetable kingdoms, are in reality only permanent varieties; and there seems to be little room for doubt that this opinion is in some instances well founded.
OF THE DETERMINATION OF SPECIES—PHENOMENA OF HYBRIDITY.
WHEN any given tribes of plants or of animals are so distinguished from each other as to render it doubtful whether they belong to one species or not, several ways have been proposed for the solution of this inquiry. The most obvious and direct one is to shew that the same difference commonly, and within ordinary experience, has arisen in the same stock to which both the tribes are referred. If that can be done, the question is at once  answered. But when the tribes about which the inquiry is made are either permanent varieties or separate species, there is greater difficulty in arriving at a determination.
In such instances there is one way of coming to a conclusion, which many naturalists prefer to adopt, and it is altogether satisfactory, if we can rely on the universality of an observation on which it is founded. I allude to the facts connected with what is termed hybridity.
Besides this criterion, there is another method of resolving the problem, but one which involves a long and often laborious research into the history of species. I shall have much to say on this subject after briefly surveying the phenomena of hybridity.
Nothing is more evident than the fact that all the tribes both of the animal and vegetable worlds are generally reproduced and perpetuated without becoming blended and mixed together. The law of nature decrees that creatures of every kind shall increase and multiply by propagating their own kind, and not any other. If we search the whole world, we shall probably not find one instance of an intermediate tribe produced between any two distinct species, ascertained to be such. If such a thing were discovered it would be a surprising anomaly. The existence of such a law as this in the economy of nature is almost self‑evident, or at least becomes evident from the most superficial and general survey of the phenomena of the living world: for if, as some have argued, there were no such principle in operation, how could the order, and at the same time the variety, of the animal and vegetable creation be preserved? If the different races of beings were intermixed in the ordinary course of things, and hybrid races were reproduced and continued without impediment, the organised world would soon present a scene of universal confusion; its various tribes would become every where blended together, and we should at length scarcely dis-cover any genuine and uncorrupted races. It may, indeed, be said that this confusion of all the living tribes would long ago have taken place. But how opposite from such a state of things is the real order of nature! The same uniform and regular reproduction of species still holds throughout the world; nor are the limits of each distinct species less accurately defined than they probably were some thousands of years ago. It is plain that the conservation of distinct tribes has been secured, and that universally and throughout all the different departments of the organic creation.
Strong as is the probable truth of this view of the economy of nature, it was long before naturalists were brought to admit the facts to be as they thus present themselves; and many vacillations of opinion may be traced among vegetable and animal physiologists on the subject. Among botanists, the most erroneous notions have prevailed. Linnaeus, whose insight into the system of nature seemed in many respects so penetrating, adopted a very singular opinion on the extent of hybrid productions among plants. He supposed them to take place between plants of different natural families. He looked, for example, on the veronica spuria as the intermediate offspring of the veronica maritinza and the verbena ojicinails; he supposed the saponaria hybrithi to be produced from the s. officinalis, fecundated by a gentiana; the actea with white fruit to be produced by the actea with black fruit, fecundated by the rhus toxicodendron. Linnaeus was ready to admit facts of this kind on mere conjecture; and when he met with a plant which resembled two others that happened to grow near to it, put it down without further evidence as their hybrid offspring. These opinions have been since regarded as wholly erroneous. Attempts to produce by art such productions between plants of different families have, as M. De Candolle  observes, uniformly failed, and they very rarely succeed between genera of the same family. Between species of the same genera, hybrids are, as it is well known, frequently produced in gardens. In the state of nature they are comparatively rare. M. De Candolle, after a critical examination of the examples which have been adduced, has drawn the following conclusion:—” Que, quoique l’attention des naturalistes soit éveillée depuis plus d’un siècle sur les hybrides, et que leur tendance ait paru être plutôt de les exagérer que de les réduire, on ne peut citer encore qu’une quarantaine d’exemples prouvés d’hybridité naturelle, et tons entre espèces de même genre, et même presque tous entre espèces de Ia même section du genre. Nous pouvons par ce fait apprécier l’hypothèse trop hardie de Linné, qui présumait que Ic nombre des espèces était allé en augmentant dune manière très marquée depuis l’origine des êtres organisés, qui avait même soupçonné que le croisement des familles avait crée les genres, et que celui des genres avait crée les espèces.” *
 * De Candolle, “Physiologie Végétale.”
But although hybrid plants are produced, there are no hybrid races. This is a fact now universally admitted among botanists. It seems that nature has prevented the perpetuation of such productions by a variety of organic defects. M. Dc Candolle conjectures that the pollen of hybrid anthers is wholly or partially deficient in granules, and that on this difference depends the absolute sterility of some, and the comparative, though still defective, fecundity of other, hybrid plants. That some cause of this description must influence the results of experiments would appear evident from the observation of M. Gaertner, who found that the number of grains fertilised in each fruit is much less in the attempts to produce hybrids than in the natural process. It has been conjectured, also, by M. De  Candolle, that abortion of the germs, or some monstrosity in the organs of fructification, is among the causes which impede the reproduction of hybrid flowers. It appears, however, that in some instances these hybrid plants can be made to reproduce, either by blending them with the primitive kinds or with other hybrids. But this rare fertility has never been known to become permanent: according to Professor Lindley, it has never exceeded the third generation. The result of all the observations which have been made upon this subject is, as M. De Candolle has remarked, that all such intermediate breeds tend incessantly to extinction, by the difficulties which are opposed to their reproduction. This explains the rarity of their appearance, and reconciles the permanence which is observed among the distinct species of nature, with the real existence, often however exaggerated, of hybrid or temporary productions, which are thus reduced into the class of monstrous and irregular phenomena in the vegetable world.*
* Mr. Knight, who has made more extensive observations on this subject than most other persons, holds most strongly the doctrine of the sterility of hybrid plants. He says, that amongst different tribes referred to the genus Prunus, the Domestica, the Inscititia, and Spinosa are likely to produce perfect offspring. He has still less doubt respecting the Armeniaca and Sibirica. The former is found in a wild state in the Oases of Africa, where it bears a rich and sweet fruit of a yellow colour: the fruit of the Sibirica is black, acid, and of small size. Nevertheless, he adds, “if these apparently distinct species will breed together, and I confidently expect they will, without giving existence to mule plants, I shall not hesitate to pronounce them of the same species, as I have done relatively to the scarlet, the pine, and the Chili strawberries.” On similar grounds be infers the specific identity of the peach and the sweet almond. If the hybrid plant is productive, one of two things will be proved; either the specific identity of the two original plants, or the transmutability of the species. But if the peach were an originally distinct species, where could it have been concealed from the Creation to the reign of Claudius Caesar? “The apple or crab of England and of Siberia,  however dissimilar in habits and character, appear,” says Mr. Knight, “to constitute a single species only, in which much variation has been effected by the influence of climate on successive generations.” The same writer states his opinion as follows in general terms:—” I have never yet seen a hybrid plant capable of affording offspring, which has been proved, with any thing like satisfactory evidence, to have sprung from two originally distinct species; and I must therefore continue to believe that no species capable of propagating offspring, either of plants or animals, now exists which did not come as such immediately from the hands of the Creator,”—in other words, that no hybrid is prolific.—Observations on Hybrids, by T. A. Knight., Esq., p. 253 of his collected Works.
 The history of hybrids in the vegetable creation has lately been made the subject of two comprehensive works by Gaertner and Wiegmann; and a comparative survey of the conclusions obtained by these writers and of all that has been established in relation to the same subject is to be found in the “Neues System der Pflanzenphysiologie” of Meyen. The following brief statement of these results is from the pen of Professor Wagner:—
“1. That hybrid plants in a natural state are very seldom produced, and that the greater number of the reputed instances rest on no sufficient evidence. 2. That hybrid plants are very seldom fruitful among themselves, but that such hybrids as the verbascum hybridum and the digitalis purpurascens from the d. purpurea and lutea, according to the corresponding observations of Koelreuter and Wiegmann, and all others which hold exactly an intermediate place between the parent plants, are absolutely barren; while those which, owing to the proportion of pollen,* partake more of either kind, and those which spring from the fertilisation of such hybrids among themselves, are occasionally propagated. 3. That plants produced from different varieties of the same species are altogether fertile, and that no impediment exists to their  propagation, while hybrids either revert to the original character, generally of the maternal parent, or become gradually less capable of reproduction, and, within a few generations, entirely extinct.”
* Mr. T. A. Knight, however, was of opinion that the proportion of pollen is a matter of indifference.
A similar law prevails in the animal creation, and its effects are, on a great scale, equally constant and uniform. Mules and other hybrid animals are produced among tribes in a state of domestication; but, except in some very rare instances occurring in particular tribes of birds, they are unknown in the wild and natural state. Even when individual hybrids are‑produced, it is found impossible to perpetuate from them a new breed. It is only by returning towards one of the parent tribes that the offspring of these animals is capable of being continued in successive generations.
It has been shewn satisfactorily by Professor Wagner that nature has established the sterility of hybrid animals by a really organic impediment. But for the full elucidation of this subject I must refer my readers to his work on physiology.*
* A succinct statement of the facts connected with the whole of this subject has been given by Professor Wagner in a supplemental note to his German translation of my “Researches into the Physical History of Mankind.”
Recapitulation and Application of the Result.
It seems to be the well‑established result of inquiries into the various tribes of organised beings, that the perpetuation of hybrids, whether of plants or animals, so as to produce new and intermediate tribes, is impossible.
Now, unless all these observations are erroneous, or capable of some explanation that has not yet been pointed out, they lead, with the strongest force of analogical reasoning, to the conclusion that a number of different tribes, such as the various races of men, must either be incapable  of intermixing their stock, and thus always fated to remain separate from each other, or, if the contrary should be the fact, that all the races to whom the remark applies are proved by it to belong to the same species.