part 4] 



BOWENENSE, aElS". ET SP. IS^OT 



291 



Diac/rammatic restoration of Nummu- 

 lospermuui. bowenense, gen. et sp. nov. 



—\ — sa. 



split readily. Tliey are approximately circular, or oval, with a 

 broadly acute apex, in which, in some cases, there is a marked 

 depression (PL XXI, fig. 3). The length of the impression varies 

 from 9 to 11 mm., and the breadth from 8 to 11 mm. 



The sarcotesta is well-developed, and is much wider in the lateral 

 and aj^ical regions than in the basal portion, where it becomes 

 comparatively narrow. Near the apex it varies from 3 to 5 mm. 

 in thickness, while near the base it is not more than 1 mm. thick. 

 Very little definite indication can be obtained from the speci- 

 mens regarding the nature of the sclerotesta. In some of them 

 there is a suggestion of a distinct narrow layer immediately outside 

 the nucellus (PI. XXI, fig. 4, sc), and this may indicate a thin 

 inner sclerotesta, outside which there is the thick sarcotesta. 



There would appear to 

 be no inner lnjer of 

 sarcotesta, as is the case 

 in some such seeds. 



The nucellus is cir- 

 cular in the principal 

 plane, its diameter being 

 about 6 mm., with a 

 di.stinct nucellar beak 

 up to 2 mm. long and 

 somewhat less than 

 1 mm. broad. This 

 nucellar beak would con- 

 tain the pollen -chamber, 

 which would appear to 

 be of comparatively 

 large size. In some of 

 the specimens it is ob- 

 vious that the nucellar 

 beak is placed opposite 

 the narrow micropyle 

 leading to the apex of 

 the seed. 



In general shape the 

 nucellus resembles that 

 of StejjJianospermtcm 

 ahenioides, as figured 

 by Coulter & Cham- 

 berlain (Bibliography, 

 No. 4, p. 43, fig. 45), and of Trigonocarpiim Farhinsonii {ihicl. 

 p. 44, fig. 46), both of which are Radiosperms. The nucellar region 

 in a number of examples is occupied by a dark, somewhat vesicular- 

 looking mass; but this does not show any preservation of structure 

 when examined under the microscope. 



_ As in many fossil seeds, the preservation of details is not suffi- 

 ciently good to enable one to determine to what extent the 

 nucellus may have been free from the integument. 



Section throug-h principal plane (X 4). mi = 

 micropyle; 7t&= nucellar beak; wf = nu- 

 cellar bundles ; pc = pollen- chamber ; sa = 

 sarcotesta; sa.i.=bundles in the sarco- 

 testa; so = sclerotesta ; v6= vascular bundle 

 entering the base of the seed. The degree 

 of attachment between nucellus and sclero- 

 testa cannot be determined from the 

 specimens. 



