1895. NOTES AND COMMENTS. 87 



for the hypothesis of hybrid origin ; that there is good evidence for 

 the origin of the cultivated form from C. cvuenta ; that Mr. Thiselton- 

 Dyer was expressing a pious opinion rather than an elaborated judg- 

 ment when he stated that " it had been accomplished by the gradual 

 accumulation of small variations." We are inclined to share his 

 belief; but we think that Mr. Bateson's polemic has been of service 

 in that it has led to distinction between inference and known fact. We 

 shall await with great interest the results of the work in progress at 

 Kew on variation. It would be of inestimable advantage to science 

 if we had the result of a series of experiments made under such com- 

 petent guidance, on the relative possibilities of raising varieties from 

 " sports," and by the selection of small variations. Perhaps the truth 

 may lie between Mr. Bateson's theory of " sports " and the more 

 orthodox view. Gradual selection may lead to the fixation of " forms 

 of organic stability," which occasionally appear as " sports." 



Fossil Plants of the Coal-Measures. 



The last of a long series of papers in which the organisation of the 

 Coal-Measure plants has been patiently worked out, and which forms 

 one of the most valuable contributions to Palaeontology issued by the 

 Royal Society, is that on Lyginodendvon and Heterangium. It is the joint 

 production of the late Professor Williamson and Dr. Scott. The 

 former, active to the last, has just left us, at a good old age ; to the 

 latter we look for a continuation of the researches to which the late 

 Professor gave so strong an impulse. The paper deals with the 

 vegetative structure of the two genera, which are described as among 

 the most interesting and, at the same time, the most puzzling repre- 

 sentatives of the Carboniferous flora. Nothing is certainly known of 

 the reproductive organs, but their morphology and anatomy point to 

 a position between ferns and cycads. The leaves of Lyginodendvon are 

 so like fern-leaves in form, venation, and minute structure, that they 

 would, if our knowledge ended here, be referred at once to Filices. 

 But the stem-structure suggests a cycad, the vascular bundles 

 exactly resembling those in the leaves of existing Cycadeas. One of 

 the authors has, however, recently found that the peculiar mode of 

 development of the bundles associated with the latter may extend to 

 stem-structures, namely, in the genus Stangevia. The foliage of Heter- 

 angium, though less well preserved, is of the same type, while the 

 primary structure of the stem recalls that of a fern like Gleichcnia ; the 

 bundles continued from the leaves, on the contrary, closely resemble 

 the foliar bundles of a cycad. In both genera secondary growth in 

 thickness occurs, but, as the authors point out, this is no argument 

 against a fern alliance, as the same happens in the modern genera, 

 Botrychiimi and Helminthostachys. In conclusion it is suggested that 

 they are derivatives of an ancient generalised race of ferns, from 

 which they have already diverged considerably in the cycadean 



