Actinaria G 111 



perforations in the body-wall, now called cinclidse, characteristic of the family 

 Sagartiadce. He placed in the genus 10 species. At least 7 of these belong to the 

 Sagartiadce. The three that he had personally studied were C. effoeta and C. 

 polypus, now in Calliactis; C. palliata, now the type of Adamsia. Others were 

 plumosa, the type of Metridium Oken; and bellis, the type of Cereus Oken. C. 

 filiformis (Rapp) is a Sagartian; C. diaphana (Rapp) is an Aiptasia, of the same 

 family. 



The affinities of the three others are different: coriacea ( = crassicornisf) 

 belongs to Urticina, in which it was included on a previous page (p. 337) by 

 him. Two other species, verrucosa and glandulosa, have been referred to Bunodes t 

 but he described both of them as having vertical rows of pores. He evidently 

 mistook the suckers, as figured, for pores. Of the verrucosa, his first species, he 

 stated that no good figure existed. It is a typical Bunodes or Tealiopsis without 

 pores. Yet McMurrich makes it the type of Cribrina, the "sieve-anemones," 

 presumably because it was the first in the list of species, thus utterly ignoring 

 Ehrenberg's diagnosis. But Ehrenberg did not always put his more typical 

 species first, but rather in the middle of his lists. At any rate it is absurd to 

 take for the type of a genus a species that is the opposite of what the author 

 intended and described, and one that evidently had a place erroneously in his 

 list only because of bad figures. 



The only real generic character that he indicated is the perforated wall. 

 Every species that he included either has perforations or else the figures, then 

 available, led him to describe them as perforated. However, Cereus of Oken, 

 1815, was based on the same feature and included some of the same species. 

 Therefore Cribrina should properly be regarded as its synonym. Were it to 

 be revived at all, it should displace Adamsia or Calliactis, for those were the 

 forms that he had personally studied and correctly described. Moreover, 

 he had already put papillosa and other verrucose species under Urticina on a 

 previous page, stating that the papilla are imperorate. 



He also put crassicornis in Urticina (first species), and that has imper- 

 forate papilla? or suckers, and may be identical with coriacea, which he placed in 

 Cribrina, misled by figures, as in the case of verrucosa. Thus, to use one of 

 these erroneous species as the type of Cribrina, as McMurrich has done, would 

 make Cribrina a synonym of his Urticina. 



Such a result shows the absurdity of entirely shifting the intentions and 

 meaning of the author as to his genus Cribrina. If used at all it should be used 

 only for perforate species of the family Sagartiadse. 1 



The species that Danielssen referred to Bunodes had acontia and verrucose 

 walls, with an endodermal sphincter muscle. It should be referred to a different 

 genus and also to a different family. Perhaps it should go in Madoniactidce 

 (Dan.), if that family is to be kept separate from Sagartiadse. That family has 

 acontia, an endodermal sphincter muscle, and verrucose or sucker-bearing 

 walls, according to Danielssen. 



The type genus, Madoniactis, he states has six sterile perfect mesenteries. 

 By extending the family to include genera with 12 or more perfect mesenteries 

 it might include all the genera having acontia and an endodermal sphincter, 

 and often verrucose walls, thus resembling typical Bunodes or Tealiopsis exter- 

 nally, when the cinclidse are not easily to be seen, and still more closely 

 resembling Actinauge and allied genera. These differ in having the sphincter 

 mesoglceal. 



Hertwig's Bunodes was a Chitonactis, one of the Sagartiadae of the sub- 

 family Chondractininae, and his family Bunodidse is a synonym of Sagartiadae. 



1 See also Haddon, op. cit., 1889, pp. 323, 324, who has independently expressed the same views. 



