640 



NA TURE 



[November i i, 1922 



We are dealing, in the Lower Carboniferous Primo- 

 filices, with early races, already specialised on their own 

 lines, and probably only indirectly connected with the 

 main current of Fern-evolution. 



The " Seed-Ferns " or Pteridosperms appear to have 

 attained a great development in Lower Carboniferous 

 times. A considerable variety of seeds is met with, 

 and in some cases there is strong evidence for attribut- 

 ing them to plants with a fern-like foliage. In one such 

 example, described by Nathorst, the seed (Thysano- 

 testa) is remarkable for having a distinct pappus ; it 

 was thus adapted to wind-dispersal, like the achenes 

 of Composites. 



No less than six families, referred to Pteridosperms, 

 are known by their anatomy. In only one is there any 

 evidence as to the seed, but all these groups show, in 

 their structure, a nearer relation to known " Seed- 

 Ferns " than to any other phylum. The case referred 

 to is that of Heterangium, a genus with a solid wood 

 and no pith. A beautifully organised seed (Sphaero- 

 stoma, Benson), obviously related to that of the Upper 

 Carboniferous Lyginopteris, is found in close association 

 with Heterangium Grievii and probably belonged to it. 

 The two genera, Heterangium and Lyginopteris, are 

 closely related, as shown by Dr. Kubart's discovery of 

 intermediate anatomical features, in species of Mill- 

 stone Grit age. 



The Lyginopteridese extend to the Upper Carbonifer- 

 ous, but the other five anatomical groups are peculiar 

 to the Lower. 1 They show a great variety in structure, 

 but none of them bear any anatomical resemblance to 

 contemporary Ferns. Our knowledge of so many, 

 more or less isolated, types indicates that we have only 

 found a few relics of what was really a most extensive 

 class of plants. 



The family most richly represented is that of which 

 Calamopitys is the type. A number of species of 

 Calamopitys are known ; they are plants with a pith 

 (sometimes " mixed "), large leaf-trace bundles, and 

 much secondary wood. The petioles, often of large 

 size and with many vascular strands, have long been 



1 Space does not admit of any account of their remarkable characters. 

 The five type-genera are: Rhetinangiurn, Stenomyelon, Protopitys, Clad- 

 oxylon and Calamopitys. 



known as Kalymma. Some of the species, with dense 

 secondary wood of a Coniferous type, have been 

 separated by Dr. Zalessky under the name Eristophyton. 

 An interesting new genus, Bilignea, in which the pith 

 is replaced by a central column of short tracheids, has 

 been discovered by Dr. Kidston. 



Apart from the " Seed-Ferns," we have the remark- 

 able Lower Carboniferous family of the Pityeae, already 

 represented, as we have seen, in the Upper Devonian. 

 Pitys was a genus of trees, with a relatively large pith 

 traversed by slender strands of wood, while the second- 

 ary wood was of an Araucarian type. The foliage was 

 quite unknown until recently, when Dr. Gordon dis- 

 covered the leaves attached to the twigs in a species 

 from the shores of the Firth of Forth. The leaves are 

 totally different both from those of any Pteridosperm 

 and from the well-known foliage of the Upper Carboni- 

 ferous Cordaiteae ; they rather resemble the needles of 

 a Fir, though more complex in structure. Dr. Gordon 

 suspects an affinity with Araucarian Conifers. 



Perhaps the chief conclusion that follows from this 

 hasty sketch of the earlier floras is the great distinct- 

 ness of the main phyla. 



The Lycopods may perhaps become merged, as we 

 trace them back, in the early Devonian Psilophytales, 

 but nowhere approach any other group. 



The Articulate appear as an isolated phylum 

 throughout. 



The Ferns may have come from thalloid plants, 

 through some of the forms of Early Devonian age, 

 where the frond is only represented by a bladeless 

 rachis. The " Seed Ferns " now appear as a totally 

 distinct line, parallel in certain respects to the true 

 Ferns, but nowhere joining them, unless it be in some 

 common thalloid source, about the Psilophytales level. 



The higher Gymnosperms, represented in the period 

 considered by Pitys and its allies, may have passed 

 through an earlier Pteridosperm stage, but this is not 

 proven. The Spermophyta generally may, for all we 

 know, be as ancient as any other vascular plants. 



Thus phylogeny still eludes us, though it remains 

 the ultimate goal of the palaeontologist. 



Obituary. 



Dr. C. G. Knott, F.R.S. 



THE sudden death of Dr. C. G. Knott, reader in 

 applied mathematics in the University of Edin- 

 burgh, and general secretary of the Royal Society of 

 Edinburgh, has deprived physical science of a devoted 

 follower and an accomplished exponent. On Wednes- 

 day, October 25, he was lecturing as usual and attend- 

 ing, in the afternoon, to the business of the Royal 

 Society. At night he was taken ill and died of heart 

 failure in a few hours. 



Born at Penicuik in 1856, Knott entered the Univer- 

 sity of Edinburgh in 1872 and soon joined a little band 

 of enthusiastic workers in the laboratory of Prof. 

 Tait. To study under that great teacher was a 

 privilege and an inspiration. The laboratory, then a 

 new feature in university physics, was a small attic, 

 meagrely equipped. Only a few of the best pupils 



NO. 2767, VOL. I ioj 



cared to seek admission ; they plunged at once into 

 research, either sharing in the investigations on which 

 Tait happened to be engaged, or undertaking some 

 independent inquiry of their own. Tait was then 

 collecting data for his thermoelectric diagram, and 

 Knott's training was to measure the electromotive 

 forces between pairs of some twenty different metals, 

 through a wide range of junction temperatures. He 

 also began the series of magnetic researches he was 

 afterwards to pursue with the help of his own Japanese 

 pupils. In 1879 he was appointed Tait's assistant, but 

 gave up that post in 1883 when he became professor 

 of physics in the University of Tokyo. After eight 

 years as professor in Japan he returned, in 1891, to 

 his own University of Edinburgh, where he spent the 

 rest of his life, at first as lecturer and later as reader 

 in applied mathematics. He also acted as the official 

 adviser of students reading for honours in mathematics 



