218 STUDIES IN FOSSIL BOTANY 



external cortex, within which the periderm had formed 

 (Fig. 89,/and^). 



If we compare the structure of Sigillaria Menardi 

 with that of the Lepidodendreae with secondary 

 thickening, such as L. Wunschianum or L. brevifolium, 

 we find one difference of importance. The primary- 

 wood in the Lepidodendreae forms a continuous zone, 

 usually of considerable thickness, while that of Sigillaria 

 Menardi is less in amount, and is broken up into 

 a ring of distinct and definite bundles. The latter 

 difference is striking enough, but, as we shall see, it 

 does not serve to distinguish Sigillaria generically. 



The inference which Brongniart himself drew from 

 the study of the structure of his Sigillaria was this : 

 " The arrangement of the woody tissue in bundles 

 composed of radial series is a character foreign to all 

 the Cryptogams ; it is, on the contrary, characteristic 

 of the Dicotyledons " ; : and hence he was led to 

 conclude, " that the Sigillariae and Stigmariae consti- 

 tuted a special family, entirely extinct, probably 

 belonging to the great division of Gymnospermous 

 Dicotyledons." 2 This conclusion, which was natural 

 enough at that time (1839), still has an historical 

 interest, owing to the dominant influence which the 

 views of this great investigator long exercised, especially 

 in his own country. The force of Brongniart's argu- 

 ment is now, of course, entirely invalidated by the 

 discovery of a great number of Cryptogams with 

 secondary growth, and by our knowledge of the 

 Cryptogamic fructification of Sigillaria itself. 



In 1875 MM. Renault and Grand'Eury described 



1 "Sigillaria elegans," p. 440. 2 I.e. p. 447. 



