character and abundant, and appear to place the matter in a very clear light, unless, indeed, we refuse to accept evidence drawn from any variations. I should state here that Professor Bower has already used certain variations (chiefly variations in the way of reduction) of the sporophylls of Tmesipteris as evidence as to the nature of the synangium. But with one partial exception he, apparently, did not -observe the variations which I propose to describe here. But the material available for his investigations is stated to have been of limited amount, and it is clearly in the native country of Tmesipteris, that the question can be investigated to the greatest advantage, for there the quantity of material available is practically unlimited, and the conditions under which the plants are growing can be best studied.
The variations now to be described may be roughly arranged in three categories
1. Sporophylls with repeated dichotomy and 2 or 3 synangia.
2. Sporophylls in which the synangium is not sessile but raised up on a stalk or pedicel.
3. Sporophylls in which the synangium is replaced by a leaf lobe of normal appearance.
1. A common variation of the sporophyll is that one of the branches forks a second time, a second synangium of normal form being carried just below the second fork. That is, a single sporophyll carries two synangia, and there are three flattened leaf lobes. A less common variation is for both of the branches of the first dichotomy to fork .again. The stalk of the leaf divides dichotomously into two branches, which are rather stalk-like in character ; each of these, after reaching .a certain length, forks again, the branches this time becoming flattened and leaf-like in form ; the whole leaf thus terminates in four leaf-lobes or leaflets, of the same character as the two leaf-lobes of the commoner type of sporophyll. There is a synangium not only at the first fork, but also at each of the forks of the second order. The three synangia are all of normal form and character, and may be of practically equal size.
It will be remembered that sporophylls and vegetative leaves commonly alternate in zones on the same shoot. Professor Bower has stated that, on the whole, abnormalities in the sporophylls and synangia commonly occur at the beginning or end of a fertile zone, .as if the abnormal condition might be referred to less perfect conditions of nutrition. Whilst I have found this to be true for variations of reduction in the leaf-lobes or synangia, it does not apply ;to these cases in which the sporophyll shows a development in excess of the normal. On the contrary, they occur rather towards the middle of a fertile zone, often many together, and with neighbouring normal
