12 



shift their position so that in individuals 5 mm long their inner openings 

 are located on the dorsal surface of each gonad sac near its posterior 

 end and close to the mid line. Before the complete establishment of 

 these relations the cells lining these canals develope a heavy ciliated 

 coat, and in those species which possess glandular enlargements (mucous 

 or albumen glands erroneously termed uteri) signs of glandular activity 

 make their appearance. It thus follows that in specimens not over 7 mm 

 long the reproductive system is definitely established with relations 

 practically as in the adult, the chief difference being the definite paired 

 nature of the gonad. 



In Nuttalociiiton hy adesi and in Acanthopleura echinata (1. c. p. 113) 

 Plate has found that the gonad even in comparatively large animals has 

 no connection with the gonoducts and since these last named tubes are 

 usually sharply differentiated into an inner glandular and an outer non 

 glandular, ciliated section he considers it probable that they have different 

 origins, the distal portion arising as an ectodermic invagination while 

 the glandular is a product of the gonad. As noted above not less than 

 twelve species of chitons examined on this point give no evidence of this 

 double origin for in its early development the gonoduct is a tube of like 

 appearance throughout, being of even calibre and ciliated. Hence I be- 

 lieve that where it is separated from the gonad until a late period in life 

 it is to be looked upon as a very highly modified condition. 



6. Note on Eremicaster, a Genus of Starfishes. 



By "Walter Iv. Fisher, Stanford University, California. 



eingeg. 11. Mai 1907. 



In 1905 (Bull. Bureau Fisheries, XXIV, p. 293) I described Ere- 

 iïiicaster as a subgenus of Porcellanaster, with tenehrarius as type. Re- 

 cently (Zool. Anz. 1907 p. 317) Prof. Ludwig raised this to the rank 

 of a genus and changed the type to Porcellanaster crassus Sladen. In 

 the genus he includes the following species: crassus Sladen, gracilis 

 Sladen, tenehrarius Fisher, and ivaliharii Ludwig. In Porcellanaster 

 s. str. are placed: caeruleus Wyv. Thom., cauli fer Sladen, tuberosus 

 Sladen, graiiulosus Perrier, inermis Perrier, pacificus Ludwig, vicinus 

 Ludwig. 



Ludwig rightly contends that the three characters mentioned by 

 me — the presence of 3 cribriform organs, segmental pits and papillae, 

 and 1 or 2 adambulacral spinelets — occur independently in the wider 

 genus Porcellanaster. The last two characters have not the importance 

 I supposed in 1905, for a species has since been described — P. paci- 

 ficus Ludwig — which has 3 cribriform organs but no segmental pa- 



