now under consideration. For it would allow the transmission of qualities from either parent, undisturbed by the influence of the other, to the offspring. In this part of his work the breeder once more finds occasion for the utmost skill and judgment ; but so dif ficult to formulate are the fruits of his experience that he often seems guided in his choice by instinct rather than by reason. Every new breed must originate in a few individuals possessing some spe cial peculiarities. Therefore, nearly-related individuals must at first be matched; in other words, close "in-and-in" breeding must be practised, or the race cannot be "fixed." In consequence of the uniformity obtained by pure breeding, characters otherwise unim portant become valuable as marks of purity of race. Thus the dark red colour of the Devon cattle becomes a criterion of "blood."
The advantages of in-and-in breeding have been insisted upon by the improvers of our domestic breeds, and some of them have declared that no ill results follow from the practice. But in spite of this assertion it is generally admitted that degeneration cither in constitution or in other ways does ultimately ensue ; so that at any cost the breeder is absolutely compelled to admit blood ^from another family or strain of the same race. In speaking of this necessity iu the case of sheep, Youatt says that the breeder will choose "a ram from a soil and kind of food not dissimilar to his own, . . . with points as much resembling his own sheep as may be quite as good as those in his own flock superior if possible in some points, inferior in none." But in opposition to Youatt it may be argued, from the practice followed by some great poultry breeders, that animals having the same physical characters, but which have been kept under different conditions, ought to be selected for crossing. By this means tone and vigour are infused into the stock without materially altering its character. In other cases a different plan has been followed. For instance, Colling (for what purpose is not clear, on account of the secrecy in which he carried out his art) crossed his short-horns with a distinct breed the Gallo way. He thus produced a sub-strain or family, called in reproach the "Alloy," but possessed of great merits, which, by recrossing with short-horns, became quite equal to the pure breed,[1] and produced animals which sold for enormous prices. This method of making one "violent" cross, and trusting to subsequent re- crossing with the pure parent form (together with long con tinued selection), has sometimes been followed where some especial quality is required. Lord Orford s well-known attempt to infuse pluck into his greyhounds by means of a cross with a bull-dog is a case in point. Stonehenge records a carefully-observed experiment of the same kind, which shows that the objectionable form of the bull-dog can be thoroughly eradicated even in four generations. In other cases a cross with a distinct variety is effected with the object of forming an intermediate race which shall transmit its characters.
Crossing.—An injudicious exaggeration of certain qualities, as in some cases before alluded to, has taken place in breeding long-woolled sheep. Here the fleece has been almost exclusively attended to, and the quality of the carcase allowed to deteriorate. No doubt, an improved breed remedying this evil might have been formed by selection but this process would have been slow and extremely difficult ; and, fortunately, there existed the readier method of forming a cross-breed race combining the desirable characteristics of both varieties. Messrs Druce and Pusey[2] have pointed out the great increase of profit yielded by a cross between the long and short- woolled sheep The following table gives in the first column the number of Cotswolds, Southdowns, and sheep of a " cross breed" intermediate between them, which a given area will support ; the second gives the total value of fleece and carcase in each case for the number of animals given in the first column:—
 | A table should appear at this position in the text. See Help:Table for formatting instructions. |
Cotswold 100 496 Southdown 120 488 Cross-breed 115 587 Crossing has, in fact, entered largely into the formation of nearly all our improved sheep.[3]
In some cases the offspring of a first cross between dis tinct species possess valuable qualities, but owing to their sterility an intermediate race cannot be formed. If, how ever, the combination is valuable the cross may be repeated at will. The breeding of mules is a familiar example of this method. In the same way cross-bred cattle, which though not sterile are yet incapable of transmitting their valuable qualities to their offspring, are bred for the butcher by a repetition of the first cross.
Some of the more important points in connection with methodical selection and the modern art of breeding have now been briefly indicated. The results obtained have seen truly astonishing. Lord Somerville graphically remarked that the modern sheep-breeder appeared to have " drawn a perfect form and then to have given it life." These extraordinary improvements have been effected almost within the last century ; and it may be objected that because selection as now practised is of modern date, the differences which characterize many races of great antiquity cannot have been produced by man. This objection, however, is not valid, for it can be shown that an unnoticed and therefore unrecorded cause of modifica tion has long been in existence. This important agent has been named " unconscious selection ;" it is illustrated by the following case. In speaking of two flocks of the New Leicester sheep possessed respectively by Messrs Buckley and Burgess, Youatt remarks that " both of their flocks have been purely bred from the original stock of Mr Bake- well for upwards of fifty years. There is not a suspicion existing in the mind of any one at all acquainted with the subject, that the owner of either of them has deviated in any one instance from the pure blood of Mr BakeweU s flock ; yet the difference between the sheep possessed by these two gentlemen is so great that they have the appear ance of being quite different varieties."[4]
Now we may feel sure that neither of these breeders intended to alter the character of his flock, he merely strove to produce the best possible New Leicester sheep, and selected those which approached his ideal most closely. Yet owing to slightlydifferentstandards of excellence having been unconsciously aimed at in the two cases, the import ant results pointed out by Youatt arose. It is an exceed ing remarkable fact, that changes so small as not to be perceived by the trained eye of the modern breeder may by accumulation produce obvious results in the short space of fifty years. And if such changes may occur unnoticed under the supervision of men keenly alive to the possibili ties of change, a far greater field for this kind of modifica tion must have been offered before any such knowledge was general. An unperceived divergence of character will arise whenever men, actuated by some vague belief in heredity, begin to select the best individuals, roughly speak ing, for reproductive purposes. Each man will uncon sciously take a standard of excellence slightly different from his neighbours, and thus his strain will imperceptibly begin to differ from theirs. Now there can be no doubt that an amount of selection sufficient for this purpose must have been practised from a very remote period. Youatt, after an examination of the passages in the Old Testament bear ing on the subject, asserts that some of the best principles of breeding were then understood. The antiquity of breed ing is also proved by certain passages in ancient Chinese encyclopaedias.
passed through stages in which they resembled the savages of the present day ; therefore it may fairly be assumed that customs which are found among lowly developed savages are of great antiquity. Now few races are more barbarous than the Australians, yet even they take pains in the breed ing of their dogs, matching the finest together and pro viding good food for the mother in order that the young may be well nurtured. From a large body of similar evidence there can be no doubt that a degree of selection sufficient
for the development of unperceived divergence has been