Numberless hypotheses have been started to account for variation. Some hold that it is as much the function of the reproductive system to produce individual differences as it is to make the child like the parents. Darwin says "the reproductive system is eminently susceptible to changes in the conditions of life; and to this system being functionally disturbed in the parents I chiefly attribute the varying or plastic condition of the offspring. The male and female sexual elements seem to be affected before that union takes place which is to form a new being. But why, because the re-productive system is disturbed this or that part should vary more or less, we are profoundly ignorant. Nevertheless we can here and there dimly catch a faint ray of light, and we may feel sure that there must be some cause for each deviation of structure however slight." It may be useless for us to speculate here upon the laws which govern variations. The fact that these exist is what the breeder has to deal with, and a most important one it is, for it is this chiefly, which makes hereditary transmission the problem which it is. His aim should ever be _to grasp and render permanent and increase so far as practicable, every variation for the better, and to reject for breeding purposes such as show a downward tendency_. That this may be done, there is abundant proof in the success which has in many instances attended the well directed efforts of intelligent breeders. A remarkable instance is furnished in the new Mauchamp-Merino sheep of Mons. Graux, which originated in a single animal, a product of the law of variation, and which by skillful breeding and selection has become an established breed of a peculiar type and possessing valuable properties. Samples of the wool of these sheep were shown at the great exhibition in London, in 1851, and attracted much attention. It was also shown at the great recent Agricultural Exhibition at Paris. A correspondent of the _Mark Lane Express_, says: