the Greensand compared loith those of existing Species. 139 



Laminarian zone at Budleigh-Salterton tliat it is impossible 

 to examine microscopically a portion of Sponge, Ecliinoderm 

 (visceral contents), or Compound Ascidian without seeing 

 several of them. They, but for their delicate nature, Avould, 

 it may therefore be assumed, have been as numerous in the 

 Haldon deposit as they are in the deep sea and in the Chalk, 

 which they might have contributed to form just as much as, if 

 not more than, the other minute organisms found in it ; for the 

 coccolith is but a Melobesian cell [Melohesia iimceUularis, mihi), 

 which, like the arborescent M. ccdcarea on the south coast of 

 Devon, may form beds of many miles in extent, entire as they 

 die or fragmental as they pass out in a comminuted state 

 from the alimentary cavities of the lower animals (e. g. As- 

 cidia arachno'idea^ Forbes) , which feed upon their protoplasmic 

 contents most voraciously. Hence, too, perhaps their pelleted 

 grouping in the form of coccospheres, if these are not their 

 sporangia. 



Besides sponge-spicules in this deposit, I have seen minute 

 bivalve shells and a* few minute Foraminifera, but no remains 

 which I could in any way identify with the calcareous spicules 

 of Echinodermata, AlcyonidfB, Gorgonidte, or Ascidi^. 



EXPLANATION OF THE PLATES. 



N.B. All the figures in these plates, excepting 7, 8, and 9, are taken 

 from the fossilized remains of sponges in the Upper Greensand of Haldon 

 Hill, near Exeter, and of Black Down, near OuUompton, respectively ; 

 and all are drawn on the scale of l-24th to l-1800th of an inch, in order 

 that their relative sizes may be seen and their nieasui'ements respec- 

 tively computed by compass and rule. 



The fractured ends of the shafts or branches are represented by a point 

 or little circle within a larger one, the former being intended for the cen- 

 tral or axial canal, and the latter for the circumference of the spicule or 

 branch respectively. 



Dotted lines indicate restored parts, where they are not intended to 

 represent the lines of one spicule behind another. 



Plate VII. 



Figs. 1 & 2. Disks of Dadylocalycites Vicaryi. 



Figs. 3-5. Disks of Dadylocalycites polycUscus. 



Fig. 6. Disk of Dactylocalycites Vicaryi, lateral view, showing the shaft. 



Fig. 7. Disk of Dactylocalyx polydiscus, Bk., after Schmidt, showing the 



ternate branching of the axial canal of the shaft (a). 

 Fig. 8. Disk of Dactylocalyx polydiscus, Bk. (Discodmnia polydiscus, 



Bocage) , after Bocage, showing the ternate division of the axial 



canal of the shaft (a). 

 Fig. 9. The same, lateral view, showing the shaft. 

 Figs. 10 & IL Rectangular branching of silicifled fibre like that oi Fii- 



j)kctella. 

 Fig. 12. Caualled silicified fibre like that of Farrca occa, Bk. 

 Figs. 13-18. Knots or branching centres of silicified libre, like that of 



Dacti/localy.r. 



10* 



