In discussions of geographical distribution in relation to problems of origin it is generally said that very nearly allied species usually occupy distinct areas, while other competent observers state the exact contrary. Lately, for example, Dr. R. G. Leavitt[28] has published an important collection of evidence upholding the latter proposition, taken chiefly from the botanical side, showing how in numerous genera two or more closely allied species coexist, frequently without intermediates, in the same localities, and may even be thus found in company throughout their distribution. The difference of opinion evidently arises from a confusion as to the sense in which the term "species" is understood and applied. Leavitt, for example, is avowedly following Jordan and, among moderns, Sargent, in applying a close analysis, and denoting as species all forms which are distinct and breed true. Against this use of the term I know no valid objection[29] but it must be obvious that if others follow a different practice confusion may result when observations are summarised in general statements. We will consider this subject again in another place, but here it may be sufficient to say that there can scarcely now be a doubt that numbers of these associated species, such as Jordan discriminated, represent various combinations of the presence and absence of Mendelian factors. This does not in any way weaken the argument which Leavitt founds upon the facts, namely, that the observed distribution of these forms is consistent with the supposition of an evolution largely discontinuous. On the other hand, those who have come to the opinion that nearly allied species generally occupy distinct ground are presumably more impressed by the characters differentiating the geographically distinct or adaptational races, seeing that genuine intermediates between them are less commonly found. Those geographical races may no doubt contain various differentiated forms; but when all live together, occasional intermediates are usually to be found even in the case of characters habitually segregating. These segregating forms Jordan would certainly have determined as species, and it must be conceded that no physiological definition has yet been drawn which consistently excludes them. FOOTNOTES: [1] Often referred to by older writers as _Meone_, Esper's name. [2] There are also two distinct island forms, unlike the European, _Xiphia_ of Madeira, and a smaller variety, _Xiphioides_ of Canary. See especially, Baker, G. T., _Trans. Ent. Soc. London_, 1891, p. 292. [3] Speyer, Adolf, and August. _Verbreitung der Schmetterlinge_, 1858, I, p. 217. [4] _Lepid. Comparée_, fsc. III, p. 372.