Il8 PAPERS FROM THE DEPARTMENT OF MARINE BIOLOGY. 



photograph. The type specimen has from ii to 12 large septa to i cm., without or 

 rarely witli intermediate small septa, septal margins flat across the collines; width 

 of valleys 3 to 5 mm. The Cocos-Keeling specmiens have precisely these characters. 

 L. tenuis (Dana) (see plate 47, figs i, la) has more acute collines and thinner and 

 more crowded septa, about 16 larger with a few smaller intermediates to i cm.; 

 L. gracilis (Dana) (see plate 46, figs. 4, 4a) has still more crowded septa up to 

 26 larger to i cm., and has broadly triangular collines, or the upper septal margins 

 may be flatly arched. Apparently there are three distmct species. The Lepturia 

 phrygia of Milne Edwards and Ilaime seems to me to be the same as tlie second of 

 the Cocos-Keeling specimens, which is described in a following note. 

 Habitat and color, Cocos-Keeling Islands. — Dr. Wood Jones states: 



"The boulders from which specimens were taken were all collected in the 'poisoned- 

 water area' of 1876 described by H. O. Forbes. Color while alive brownish, rapidly turns 

 white on exposure." 



The following notes from the same author relate to another specimen (plate 45, 

 fig- S' plate 46, fig. 3) from Cocos-Keeling: 



"Pulu rikus, rocky lagoon shore near east end; water 3 feet deep at low tide. Color, 

 bright yellow brown while alive. Four specimens from a colony about 18 inches in diameter, 

 upper surface lumpy and lobulate; base irregularly circular." 



This coral seems to me to be the one Mdne Edwards and Haime (and perhaps 

 also Dana) referred to L. phrygia, as in places there are about 15 larger septa with 

 alternating smaller to i cm., but this septal arrangement is not invariable, as 

 frequently there are no smaller between the larger septa. Usually the septal mar- 

 gins over the colhnes are truncate, but in places they form low triangles. There is 

 in places a considerable development of spongy tissue along the valley bottoms, but 

 a lamellar columella is distinguishable in the axis. The specimen resembles L. 

 gracilis in its growth-form and crowded septa, but continued comparison of it with 

 typical L. phrygia has convinced me that it is a variant of the latter species and 

 that it is not L. gracilis. 



Distribution. — Cocos-Keeling Islands; southern Philippines (J. B. Steere); 

 Cej'lon' (Dana). Reliable, precise locality records are few. 



Leptoria gracilis (Dana). 



Plate 46, figures 4, 4rt, Dana's type uf Meandrina gracilis. Also plate 17, figure 34, of Dr. Mayer's article. 



1846. Meandrina gracilis Dana, U S. Expl. Exped., Zooph., p. 261, plate 14, figs. 6, 6a, Cib. 

 1857. Leptoria gracilis Milne Edwards and Haime, Hist. nat. Corall., vol. 2, p. 407. 



Identification is based on Dana's type from the Fiji Islands, No. 16, U. S. 

 National Museum. 



Stations, Murray Island. — Southeast reef, line I : 



1,600 feet from shore, water 10 inches deep. 



1,640 feet from shore, water about 14 inches deep, bottom of broken corals. 



There are 3 specimens from Murray Island which show no marked variation 

 from Dana's type. Width of valleys 3 to 4 mm., interserial walls thin or thickened 

 up to 2 mm., on the periphery up to nearly 4 mm. Septa about 19 to the centimeter; 

 usually equal, sometimes alternating in size; upper margins rounded or flattened; 

 inner edges perpendicular. Septa much more crowded than in L. phrygia and L. 

 tenuis. 



Distribution. — Fiji Islands; Murray Island; Indian Ocean; Red Sea. 



'This is named Mieandra rudis by Verrill, Trans. Conn. Acad. Arts and Sci., vol. 11, p. 69, 1902. 



