Anatomie. 517 



At a higher level the central tracheides are replaced by a 

 parenchymatös pith. The first leaf-traces depart without 

 breaking the continuity of the xylem, although a pith may have 

 already appeared in its centre. 



The anatomy of the mature plant was studied in Todea 

 barbara, T. superba and T. hymenophylloides, and the structure 

 of the stein stele proved to be similar in all essentials to that 

 of Osmunda regalis, although the number of the medullary 

 rays is smaller and the xylem ring is more continuous; parti- 

 cularly so in T. superba. As a result of their investigations 

 the authors conclude: 1. That the endodermis is not of common 

 origin with the tissues lying internal to it; 2. That the phloem 

 is developed centrifugally ; i. e. the tangentially elongated 

 elements at the periphery of the phloem do not represent 

 protophloem but are of later origin than the sieve-tubes lying 

 within them. 



The phloem on the abaxial surface of the leaf-trace is 

 centripetally developed and its peripheral protophloem is conti- 

 nuous with the above-mentioned tangentially elongated elements 

 of the stem. 



Isolated islands of sclerenchyma are occasionally met with 

 in the medulla of Todea barbara and T. hymenophylloides, and 

 an internal endodermis was discovered in the latter which is, 

 however, very irregulär and sporadic in its appearance. No 

 indication of this internal endodermis is to be found in the 

 stem of the seedling. 



A section is given to the enumeration and discussion of 

 the various fossil impressions of sporangia and fronds resembling 

 those of the Osmundaceae. Finally, in summing up their 

 results, the authors decide that the medullary tissues are not 

 homologous with the cortical tissues, but they form part of the 

 stele, and that the internal endodermis in Todea hymeno- 

 phylloides is a specially modified part of the stelar conjunctive 

 tissue. The stele of Lyginodendron is believed to be the type 

 of structure most nearly allied to that of recent Osmundaceae. 



D. J. Gwynne-Vaughan. 



VuiLLEMlN, P., Evolution, anatomie et biologie du 

 Gentiana ciliata. (Bull. Soc. des Sc. de Nancy. Ser. III. 

 Annee III. 1902.) 



La souche florifere du Gentiana ciliata se produit par 

 gemmiparite d'une racine horizontale, depourvue de poils et 

 d'origine encore inconnue. Elle porte des racines adventives 

 soit groupees sur le bourrelet de sa base endogene, soit 

 dispersees sur toute sa longueur. Cette souche porte en outre 

 des bourgeons de reserve, isoles dans l'aisselle de feuilles 

 ecailleuses, Caches sous terre et qui assurent sa perennite; eile 

 est donc vivace ainsi que l'admettaient les anciens botanistes. 

 Les racines renferment toujours des micorhizes dans leur 

 parenchyme cortical (autoderme). Elles pre'sentent en outre 



