should get species of a sub-genus and genus,--as varieties of merino-sheep,--varieties of British and Indian cattle. Fresh species might go on forming and others become extinct and all might become extinct, and then we should have <an> extinct genus; a case formerly mentioned, of which numerous cases occur in Palaeontology. But more often the same advantages which caused the new species to spread and become modified into several species would favour some of the species being preserved: and if two of the species, considerably different, each gave rise to group of new species, you would have two genera; the same thing will go on. We may look at case in other way, looking to future. According to mere chance every existing species may generate another, but if any species, A, in changing gets an advantage and that advantage (whatever it may be, intellect, &c., &c., or some particular structure or constitution) is inherited{140}, A will be the progenitor of several genera or even families in the hard struggle of nature. A will go on beating out other forms, it might come that A would people earth,--we may now not have one descendant on our globe of the one or several original creations{141}. External conditions air, earth, water being same{142} on globe, and the communication not being perfect, organisms of widely different descent might become adapted to the same end and then we should have cases of analogy{143}, [they might even tend to become numerically representative]. From this often happening each of the great divisions of nature would have their representative eminently adapted to earth, to <air>{144}, to water, and to these in <illegible> and then these great divisions would show numerical relations in their classification. {135} After "organs" is inserted, apparently as an afterthought:--"no, and instance metamorphosis, afterwards explicable." {136} For analogical resemblances see _Origin_, Ed. i. p. 427, vi. p. 582. {137} "Practically when naturalists are at work, they do not trouble themselves about the physiological value of the characters.... If they find a character nearly uniform, ... they use it as one of high value," _Origin_, Ed. i. p. 417, vi. p. 573. {138} "We are cautioned ... not to class two varieties of the pine-apple together, merely because their fruit, though the most important part, happens to be nearly identical," _Origin_, Ed. i. p. 423, vi. p. 579. {139} The whole of this passage is obscure, but the text is quite clear, except for one illegible word. {140} <The exact position of the following passage is uncertain:> "just as it is not likely every present breed of fancy birds and cattle will propagate, only some of the best."