VARIATIONS OK KUNG1.

55

Tricholomq, from one of the subgemis Cliiocybe, because, thoughone is distinguished by having the gills emarginate or sinuatedbehind before their attachment to the stem takes place, andthe other has the gills acutely adnate without any emargina-tion, modifications occur on either side; while in Clitocybe,in an early stage, there may be a decided emargination, inTricholonia, from the depression of the pileus, the gills maybecome decurrent. And yet these characters are founded innature, and are satisfactory enough when the variations towhich they are subject are properly appreciated. Still more,changes of outward form may occasionally take place, incon-sistent with the character of the species. Thus we may haveumbonate individuals where the pileus ought simply to beobtuse, while a steraless Agaric may exhibit a stem or thecontrary. The fact, however, is, that as in phsenogamicbotany the sum of characters must be looked to, while it isremembered that no definitions in natural history can bestrictly mathematical. Where species are very difficult todistinguish, it is in general because forms are separated whicharc too closely allied, an evil which is familiar enough to everypractical botanist, though apt to be overlooked or completelyignored by the inexperienced or mere localists. The essentialcharacters are often the least superficial, and hence the youngbotanist is apt to make mistakes, from confounding mere ana-logies with affinities. Some Agarics of the subgenus Pratella,for instance, would never be separated from others of thesubgenus Lepiota, without examining the nature of the spores.These organs, moreover, sometimes differ in closely allied3pecies of such similar external characters, that it would beimpossible to distinguish them without having recourse to themicroscope.

If there is difficulty about species, there is often far moreabout genera. The characters in so natural a group are