340 Dr. J. W. Gregory — An Egyptian MUIeporoid Coral. 



canal system, wliicli ramifies through the broad areas of porous, 

 vesicular tissue that separates the gastrozooids. The massive 

 growth, although a character of little importance, helps to separate 

 the fossil from the Stylasterid^. The only remaining available 

 groups in which the fossil may find a home are the Milleporidee 

 and the milleporoid section of the Stromatopoi'idce. To which 

 of these it should be referred is not very easy to decide. If we 

 abstract from Moseley's ^ diagnosis of the MilleporidEe all the 

 characters which depend on the hard parts, we find that he states 

 them as follows : — The coenosteum is irregular, arborescent or 

 incrusting; it is composed of a thin, superficial living layer, lying 

 over dead, earlier-formed layers. The pores have no styles, and are 

 divided into vertical series of chambers by tabulas; the dactylo- 

 zooids are usually irregularly arranged, but a circle of them may 

 occur around a gastrozooid. 



Thei'e is nothing in this series of characters to exclude the 

 Egyptian fossil from the Milleporidae. It differs, however, from 

 MiUepora in two ways : the pores are less unequal in size, and 

 a transverse section shows none of the broad open cavities of 

 the gastrozooids. In the second place, the walls around the 

 interspaces in the corals are more compact and narrow. The 

 specimen agrees more nearly with the milleporoid group of 

 Stromatoporidee owing to the presence of the long, vertical pillars, 

 the linear series of pores, separated by branching grooves, and the 

 more equal size of the pores. In the presence of occasional cyclo- 

 systems it agrees, however, again with the Milleporidee. 



Among the Stromatoporidas, it agrees most closely with the 

 Idiostromid^, and especially with the genus Hermatostroma, which 

 has no axial tube, and in which ■we see the beginning of the 

 reduction of the concentric, horizontal laminEe, that are so 

 conspicuous in the typical Stromatoporoids. 



I therefore regard the fossil as one of the missing links between 

 the Palceozoic Milleporoid Stromatoporoids and the Cainozoic 

 Milleporidee, and therefore propose for it the name Millestroma. 

 The species is named after the author whose work has placed our 

 knowledge of the Palaeozoic Hydrocorallines on a sound basis. 



Family MILLESTEOMID^. 



Characters. — Coenosteum massive, encrusting. Definite zooidal 

 tubes present. The coenosteum is composed of reticular tissue with 

 abimdant tabulge, but otherwise no well-developed horizontal lamina3. 

 Micropores (dactylopores) usually irregular, linear, or in cyclo- 

 systems. No principal axial tube. 



Genus Millestroma, nov. 



Diagnosis. — Coenosteum massive ; skeletal framework irregular, 

 but forming a series of branching, anastomosing tubes. The. 

 horizontal or concentric laminse are rudimentary, but tabulse ai-e 



1 H. N. Moseley, " Eeport on certain Hydroid, Alcyonarian, and Madreporarian 

 Corals" : Eep. Chall. Exped., ZooL, vol. ii, pt. 7, p. 92. 



