378 On the Morp1u)logij of some Cretaceous Cirri'pedcs. 



tlic valves, there is much lesciuhlance to tlic forms of 

 Scaipcllum iuchidcd in the subgenus Arcoscalpellum. This 

 is more cspreially miukcd in tlie case of the n])])cr whorl 

 of valves uhich .-ij^rcc almost exactly in disposiiion, and 

 except for the more simj)le carina, uhich is not divided otf 

 into tectum, parictes, and intrapaiietes, there are no dis- 

 tinguishing features. AVhile there is a somewhat similar 

 disposition of the lower whorl of valves to that in Arcoscal- 

 pellum there is a great diti'crence structurally, for the 

 inframcdian lafus overlaps the rostral and carinal latus on 

 either side ; the loMcr laicr:il plates are hardly at all 

 specialized, for the rostral and inframcdian latus are almost 

 exactly ahke in siiape and size, and resemble very much the 

 carinal latus, which, however, is about twice as large as the 

 rostral and inframcdian latus ; the rostrum is comparatively 

 very much greater in size than in any known Arcoscalpellid, 

 in which latter it becomes exceedingly small or is entirely 

 absent in the more recent forms. CreliscaJpellunt is clearly 

 related to ArcosciilpelUim, but, \\hile the valves have com- 

 bined to form a somewhat similar capitulura, the individual 

 valves, especially the carina and the lower latera, still retain 

 their piimitive structure, and have not become so specialized 

 in shajjcas in Arcosciilpclluiii and the more specialized forms 

 of ^calpellion, s. str. 



There would seem to be some relationship also to Calantica 

 (Scillce/epas), especially in the lower whorl of valves. In 

 Scillalepas there is no upper latus interposed between the 

 scutum and tergum, the valve which is homologous with 

 the upper latus in other forms being still a member of the 

 lower whorl. The valves of the lower whorl in Scillalepas 

 are not so dilTerentiated in shape as in Arcoscalpellum^ 

 although more so than are the valves of Cretiscalpellwn, and 

 the median latus does not overlap the rostral and carinal 

 hitus, but on the contrary is overlapped on each side by 

 those valves. 



The structural resemblance of Cretiscalpellinn is therefore 

 closest to Arcoscalpellimi, 'and this })robably indicates not 

 that Arcoscalpellum was derived from L'retiscalpellmn, but 

 that they had a common ancestor, which in some way was 

 related to Scillcelepas. 



Arcoscalpellum has a known range from the Lower Cre- 

 taceous (Aptian) to Recent, Cretisculpellum is known to 

 exist still earlier in the Lower Cretaceous, (or it occurs in 

 the Neocomian, but it is not known more recent than the 

 Danian, and Scillcelepas has a known range from the Upper 

 Cretaceous (Upper Senonian) to Recent, although it is 

 probable that it existed in the Jurassic rocks. 



