138 Transactions of the Society. 



Brongniart. The evidence for this conclusion is fully stated in 

 the " Cours de Botanique Fossile " (Benault, 1883), where it is 

 shown that the type of petiole previously known as Myeloxylon 

 Landriotii, was characteristic of Alethopteris, while the structure 

 of M. radiatum belonged to species of Neuropteris, and perhaps 

 also of Odontopteris, results which confirmed the acute suggestions 

 already made by M. Grand'Eury. 



Some years previously Eenault had prepared the way by a 

 detailed study of the structure of the petioles in question (Eenault, 

 1874) which he then re-named Myelopteris. The reason for this 

 change in nomenclature was his conviction that these organs 

 belonged to Marattiaceous Ferns, an opinion which he still main- 

 tained in 1883, though by that time evidence had already been 

 brought forward in Germany to show that Myeloxylon was the 

 petiole of Medidlosa, a genus of stems then regarded as Cycadean. 

 We now know that the Neuropteridese or Medullosese were seed- 

 bearing plants, approaching the Cycads, but retaining characters 

 indicative of affinity with Ferns in a wide sense. Benault's idea 

 of Marattiaceous relationship, in fact, is still tenable, for the latest 

 view, expressed by a high authority, is that the Pteridosperms 

 (including Xeuropteridese) had a common origin with the Maratti- 

 acea?. Almost at the close of his life, Renault recognised the new 

 position which the question had assumed, for in his last published 

 work, presented to the Academy on May 15, 1904, in maintaining 

 the multiple origin of Phanerogams from different Cryptogamic 

 phyla, he cites the Medullosese, and especially Colpoxylon aeduense 

 (Eenault, 1896), the structure of which he had himself revealed, as 

 probably connecting the Ferns with the Cycads (Eenault, 1904). 



The plants which Eenault, by his anatomical researches, so 

 largely contributed to reconstruct, constituted a most remarkable 

 group. The often massive stems, with a complexity of structure 

 scarcely to be paralleled among recent plants, bore enormous 

 petioles, often 6 inches in diameter, branching out into great com- 

 pound leaves suggesting some huge Angiopteris or Osmunda ; on 

 the fertile pinnse of these fern-like leaves were borne the seeds, 

 resembling those of Cycads, and sometimes, as it appears, of an 

 immense size. The strange and unfamiliar appearance presented 

 by a female Cycas in fruit, can give us but a faint idea of the 

 bizarre habit of the extinct family. 



4. CORDAITE^E AND POROXYLEiE. 



The wonderful reconstruction of the Cordaitese by Eenault and 

 Grand'Eury, the former working from anatomical characters, the 

 latter from external features, has already been referred to, and is 

 familiar to all students of Fossil Botany. It was their work which, 



