1. The spike or strobilus occasionally branches ; perhaps one stro- bilus in two thousand will be found forked, the two divisions becoming equally developed. I am of course only speaking of the form which grows in New Zealand, and this may possibly be a slightly more robust form than that found in Australia. The branching always takes place above the lowest sporophyll, sometimes quite at the base of the spike, near the lowest leaf, sometimes further up, or even close to the apex of the strobilus.
But even when the strobilus forks there is no transition of form between the sporophyll and protophyll. I have occasionally observed on the peduncle a leaf some distance below the rest of the strobilus, but such a leaf has always been of the sporophyll type. In the Australian form, investigated respectively by Bower and Bertrand, to whom we are indebted for most of our knowledge of Phylloglossum, eight was the largest number of protophylls found on a plant, whilst Bertrand urges on anatomical grounds that six is the normal number of protophylls. I have collected plants with twenty protophylls, whilst others with ten to fifteen such leaves are of common occurrence. But even in plants richest in protophylls no transition occurs between protophylls and sporophylls. So far as any evidence here available goes, it would almost seem as if the two structures were not strictly homologous.
To express my meaning in the language of a modern theory the protophylls may have arisen from the differentiation of the lower region of a sporogonium (or the homologue of a sporogonium) in which this region had already acquired sterilised tissues, whilst the sporophylls arose from the upper fertile region of the sporogonium. If so, the protophylls cannot be regarded as sterilised sporophylls.
There appears to be no necessary connection between the number of protophylls and the reproduction by spores. Plants with two proto- phylls only may produce a weak spike, whilst plants of twenty proto- phylls may be barren.
2. In barren plants the new tuber is formed by the lowering of the apex of the stem, but in fertile plants a new outgrowth is formed, which Bower regards as adventitious. This may doubtless be con- sidered as a form of branching. Neither Bertrand nor Bower observed more than a single new tuber formed in the examples at their disposal. Bower, indeed, was inclined to infer that as no other mode of vegetative reproduction was known, the plant depended for its multiplication solely upon the germination of the spores. But I have found that the formation of two new tubers is quite a common occurrence, though plants which form a single tuber are still in the majority. The two new tubers may be formed on opposite sides of the plant, in which case a slight dispersion of the plants takes place. Sometimes the two tubers arise close together. Apparently they may be formed almost simul-
