accidentally sown in the same district, in situations that proved favourable to the one, and unfavourable to the other, than has taken place in regard to any other kind of grain cultivated here; hence it has happened, that the peculiarities of some of the different kinds of oats have been more precisely ascertained than those of other kinds of corn. Without specifying here the other varieties, I shall barely mention two sorts which differ very materially from each other. One of these yields a small thin skinned grain, which weighs very well, and yields a good quantity of meal when thoroughly ripened; it is moderately early, that is to say, it is not so late as some kinds, nor so early as others; it affords a long straw, moderately strong, and will yield a tolerable crop upon land of a very poor quality; but, if it be sown upon rich land, it grows with so much luxuriance, as to run a great risk of lodging and being rotted entirely before it be ripe, should the season prove moist in any degree. This variety is known by the name of the Blainzlie oat, from a farm of that name in Berwickshire, where it has been cultivated, without admixture of any other sort, for time immemorial, and from whence it has been gradually diffused over most parts of Scotland.

The other kind of oat ripens at least a fortnight earlier than the former, if sown at the same time in the proper season. It is a plumper and more compact looking grain, and rather more weighty, though the difference in this respect is lefs than the appearance would indicate. It requires a very rich soil to rear it to the utmost perfection; and the stalk is so strong, and at the same time so short, that it is in no dan

ger of lodging even upon the richest soil, let the season be what it may; on this account, and because it does not tiller much, double the quantity of seed of this sort may be safely sown than, durst be ventured of the other. The ear is short when compared with all other kinds of oat that I have seen, and the footstalks are stiff, and stand out from the stalks nearly at right angles. From the circumstances here stated, it generally happens that upon very rich land nearly double the quantity of grain could be reaped, if it were sown with Poland oats, for so they are usually called (the name, however, should not be relied on, as other sorts are known by the same name), than could have been obtained from the same land, had it been sown with Blainzlie oats. [There is one instance of twenty-one bolls, three firlots of Poland oats, being reaped from one acre, which is equal to one hundred and thirty bushels Winchester measure; but this is a solitary case; sixteen bolls of this kind of corn (ninetysix bushels) is no uncommon crop. The Scotch acre is to the English as five to four nearly.] On the other hand, were these kinds of corn sown both at the same time upon poor land, the case would be just reversed; for the crop of grain from the Blainzlie seed would exceed the Polish oat in the proportion of two to one nearly, and the straw would exceed the other in the ratio of four to one at least.

The practical inferences that may be drawn from these facts are obvious; but they are so important, as to deserve to be dwelt upon a little. Should it have so happened that one of these kinds of oat only was known in a particular district, and the other sort only

in another district, it is plain that the farmer in one district must have adopted a practice in regard to the culture of that kind of corn that would have been totally incompatible with that which was found from experience to succeed best in the other. They might thus justly, as they thought, have accused each other of bad management, or even of stating facts falsely; but what is of more importance to be adverted to is, that in neither of these districts could the farmer have derived nearly the same profit from the culture of this crop that he might have done; because, in either case, some part of his land must have yielded a very deficient crop, compared to what it might have done by a judicious adaptation of the crop to the state of his fields.

As the same variations doubtlefs take place in regard to the varieties of all other kinds of corn, and as these variations are diversified in regard to a vast number of particulars, it must follow, by the same mode of reasoning, that, until all these varieties are fully known, and their specific peculiarities distinctly ascertained, the farmer, to continue our former allusion, must continue to work upon materials that are, in almost every case, much worse adapted for his purpose than others would be which he might attain were they known to him; and thus the general produce of the country must suffer a prodigious abatement from what it might have been, without costing one shilling of additional expense. To ascertain facts of this nature is, therefore, an object of great importance, and falls exactly within the province of an experimental

farm.

But the diversity in regard to animals is perhaps still greater than in vegetables, and the distinctive peculiarities of the varieties of each species are certainly lefs known. This subject, though of infinite consequence to the advancement of agriculture, has indeed scarcely begun to be studied; or, where it has been at all adverted to, it has been under the influence of a train of ideas deduced from false principles, or illfounded theories of philosophers and naturalists, which serve only to lead to perplexity or error; incontrovertible facts being continually at variance with opinions that have been too hastily admitted as axioms that require no demonstration.

Having lately had occasion to take notice of this circumstance under the head of Natural-history, Vol. I. page 51, there is no occasion to recur to it in this place. It is only necefsary here to observe that, before a judicious set of experiments can be made on the subject, it will be necefsary first to collect together the facts referable to this head that are already known. As a beginning, I shall enumerate a few of the varieties of our most common domesticated animals in as far as these are yet known, adding a few particulars respecting the canine species to those already specified, not so much because Mr. Buffon has chosen it as the particular clafs by which to illustrate his theory, as that the varieties belonging to it are more numerous, perhaps more distinctly characterised, and more generally known by man than any other we are acquainted with.

VARIETIES OF THE DOG SPECIES.

THE first particular I shall take notice of as tending to prove that there is a permanent diversity in the varieties of dogs, and that they must have been derived from original distinct parents, and not all sprung from the shepherd's dog, as Buffon contends, is the difference of size. Between the smallest lap dog, not much larger than a rat, and the largest mastiff, the difference in point of weight is not lefs, I should suppose, than as one hundred to one; now it is well known that these two breeds, if carefully guarded from an intermixture with other varieties, can be preserved entire in this climate for ages, each kind without any sensible deviation. We know that there was a favourite breed of dogs nourished by king Charles the Second, which have been propagated ever since under the name of king Charles's dogs, without any sensible deviation in point of size or other circumstances, as can be clearly ascertained by the portraits of these dogs well preserved, compared with the living descendants of that breed every day to be met with; we know also that the large dog, painted by Vandyke in the picture of Charles the First's children, is to be found in all respects the same in the present day as at those times. These two kinds of dogs, if placed together without any intermediate link by which a mongrel breed could be produced, must have remained perhaps for ever distinct, from the mere physical ob