no Caxaries AXD (^age-Birds. 



with almost absolute certainty to produce nothing but five-toed birds The first pair 



have scarcely any tendency that can be relied upon to produce the desired five toes ; the other 

 pair can be depended upon as regards nearly every one. The first pair presents nothing to a 

 breeder save the foundation upon which he may, by care and perseverance, found a structure 

 hereafter; the other represents work fully done, and a 'strain^ which, as regards the one point we 

 have considered, is perfected and established, and only needs ordinary care to preserve in the 

 same perfection for an unlimited length of time. 



" We have- selected one feature as an example ; but to any other the same reasoning would 

 apply. Single or double combs in fowls ; colour or carriage of tail, or of ears, or of any other 

 point in a dog ; speed or endurance in a horse " — marking, crest, spangling, or other feature in a 

 Canar}' — "all are subject to the same laws, and can be 'fixed' in the same manner. But it will 

 readily occur to most of our readers that every animal is bred for many points, and not solely for one, 

 such as we have been considering, and that here the difficulty in breeding successfully begins, and 

 the inexperienced breeder usually finds that, as he attempts to deal with any one point, he is very 

 apt to deteriorate in some other previously attained. 



"The chief reason of this is, that the faults as well as the good points of any parent tend to 

 be perpetuated. When, therefore, it is considered that it is almost impossible to tell when all 

 tendency to revert to the features of any particular animal in a pedigree shall for practical purposes 

 be lost, the complication of the problem becomes apparent. At each step in the process of 

 breeding towards some given point, the parents have to be chosen in reference to it ; and in each 

 such case the parents introduce tendencies to produce other points which are not wanted. Nay, 

 not only do they introduce tendencies which can be known or surmised, but it will be evident at 

 once that unless their own pedigree and course of breeding are known for generations back, they 

 must introduce tendencies which, not appearing in themselves, are not known. When, therefore, 

 we consider the changeful and capricious manner in which most amateurs — in the first instance at 

 all events — conduct their breeding, we shall cease to wonder at the anomalous nature of the results 

 they often obtain." 



Continuing the theme, and still drawing his illustrations from his poultry-yard, the following 

 graphic description of a mode of breeding too common in the Canary Fancy is much to the 

 point : — " The first year he breeds, while various faults can be easily enough found amongst his 

 various chickens, he finds probably some one fault peculiarly general : it may be want of 

 leg-feather, or streakiness, or light breast in his pullets — let us suppose that it is want of 

 leg-feather. To correct this he next season buys, or selects from his own stock, a hocked bird. 

 This time he gets plenty of feather, but if his pencilling was good before it is very likely worse now. 

 So for next season he selects a bird with beautiful dark but speckled breast, and splendid hackles, 

 and he finds his pencilling somewhat improved (though not so much as he hoped), but very likely 

 his cockerels are now very light on the breast, and ten to one the old fault of want of feather 

 re-appears. He thinks now that what he wants is a fine jet-black-breasted cock, and he gets one 

 just to suit, when some of his cockerels are splendid in colour : but perhaps the father was coarse 

 in the comb, and so all the chickens are, and very probably the pencilling of nearly all his pullets 

 is quite dull and cloudy, those which are not being nearly white-breasted. We think this is a pretty 

 fair picture of average breeding. Such a plan necessarily fails in producing uniformity, simply 

 because no point is bred for long and persistently enough to fix it at all. Each time a fault is 

 attempted to be corrected some influence upon that fault is really exerted ; and, if followed up, the 

 ground might be secured ; but very little really is gained for the first year or two, and by dropping 

 the next link in the succession all, or nearly all, is lost again. Some breeders never get beyond this, 



