A MONOGRAPH OF THE EXISTING CRINOIDS 343 



But in spite of its approach to some of the genera of the Colobometridae, Amphi- 

 metra agrees better in the sum total of all its characters with the more generalized 

 species of Heterometra than it does with any of the genera of Colobometridae. 



It should be borne in mind that, except only in the case of the enlarged proximal 

 pinnules in the smallest forms, the features characteristic of the Colobometridae as 

 contrasted with the Himerometridae the occurrence of IIBr 2 series and the lateral 

 broadening of the dorsal processes on the cirrus segments are found only very excep- 

 tionally in occasional individuals of the largest form of the largest species (A. tessellata 

 tessellata) . 



History. The first notice of a species of this genus was the publication by 

 Guerin-Meneville in 1828-1837 of a figure identified as Comatula carinata (= A. 

 tessellata discoidea). 



Prof. Johannes Miiller in 1841 published the first description of a species of this 

 genus, calling it Alecto tessellata. In 1846 he described two additional species, calling 

 them Comatula (Alecto) milberti and Comatula jacquinoti. 



Dujardin and Hupe in 1862 redescribed Miiller 's three species under the generic 

 name Comatula. In 1865 Prof. Sir Wyville Thomson referred to one of Miiller's 

 species (tessellata) under the generic name of Antedon, and in 1869 Count Pourtales 

 referred to another (milberti) under the generic combination Antedon (Comatula). 

 All subsequent authors up to 1907 have assigned the species of Amphimetra to Antedon. 



Dr. P. H. Carpenter in the, Challenger report on the comatulids published in 1888 

 established a special group called the Milberti group to include species of Antedon in 

 which the pinnules of the first pair are comparatively small with their component 

 segments but little longer than broad, and one or more of the second, third, or fourth 

 pairs are more massive with stouter segments than their successors. 



To the Milberti group he assigned the following 14 species: pinniformis (= Am- 

 phimetra pinni/ormis) , serripinna ( Oligometra serripinna), carpenteri (= Oligometra 

 carpenteri), pumila (= Compsometra loveni), milberti ( = Amphimetra tessellata), 

 laemssima (= Amphimetra molleri + Decametra, sp.), tessellata (= Amphimetra 

 tessellata), perspinosa (= Colobometra perspinosa), anceps (= Heterometra quindupli- 

 cava, 10-armed individuals), variipinna (= Heterometra crenulata, 10-armed individu- 

 als), carinata (= Tropiometra carinata), parvicirra (= Dorometra parvicirra), informis 

 (= Decametra informis), and loveni (= Colobometra perspinosa). He remarked that 

 tessellata was the only comatulid of which he had not seen the type specimen. 



Carpenter wrote that this is a somewhat heterogeneous group, and he had had 

 considerable trouble in working out an arrangement that he could regard as even 

 approximately satisfactory. He noted that the definition given would almost include 

 such forms as (Isometra) angustipinna and (Thysanometra) tenuicirra, which he 

 assigned to the Tenella group (see Part 5), while (Dorometra) parvicirra, which he 

 placed in the Milberti group, though with some doubt, has many points of resemblance 

 with Antedon rosacea (= bifida) and A. dubeni. He said that milberti (= tessellata; 

 but most of the specimens he had examined were molleri) itself exhibits traces of the 

 wall-sidedness of the IBr series and lower brachials which is so marked in the Basicurva 

 group (see Part 4b), while (Tropiometra) carinata differs in many respects from 

 (Oligometra) serripinna, milberti, and the typical members of the group, so that another 

 group may have to be established for it at some future time. 



