Br. F. A. Bather — Tube-building Fossil Annelides. 555 



branched, although an artist would not need much imagination to 

 represent them as having done so. Saporta both describes and figures 

 his species as " pourvus c_a et la d'une ramification solitaire ", and 

 this, if well founded, constitutes a clear difference. 



The tubes are made up of convexly rounded pellets, brownish in 

 colour, and apparently composed of the minute brownish sand-grains 

 that may be seen with a lens scattered throughout the fine bluish- 

 grey clay of the matrix. The pellets are usually elliptical, with their 

 long axis more or less ti'ansverse to the tube. 



I do not hesitate to regard these tubes as having been made by such 

 a Tubicolous Annelid as Terebella figsulm or a Salella ; but the 

 question of a name gives rise to much hesitation. 



Phymatoderma ccelatum, which they so markedly resemble, " s'ecarte 

 tres-sensiblement, selon nous [Saporta], de toutes les Algues fossiles 

 signalees jusqu'a present [1873]." This one can understand if it is 

 not an alga but an annelid. But then it would not be a Phymatoderma, 

 which appears to have been based by Brongniart on the Liassic 

 Algacites gramdatus Schlotheim. P. ccelatum was founded on a 

 specimen from the Oxfordian of Bouches-du-Rhone. 



The genus Granularia was established by Pomel (1847) . with 

 G. repanda of Corallian age as genotype. References and details will 

 be found in Saporta (1873, op. cit., p. 108), whose figure of the genotype 

 (pi. xii, fig. la) shows an obviously branching form. There seems no 

 reason why we should not speak of Granularia ccelata (Sap.) ; but our 

 Albian form, though closely resembling that Oxfordian species, is not 

 likely to have been identical. Therefore we may speak of it for the 

 present as Granularia cf. ccelata (Sap.). 



To avoid any confusion of the nomenclature, I wish to say quite 

 definitely that I use the names Kechia, Granularia, etc., only on the 

 supposition that their genotypes, and consequently the genera them- 

 selves, are of animal origin. Should those species, with their genera, 

 ultimately be retained among plants, and should the British 

 Cretaceous species here discussed be accepted as Annelida, then no 

 use by me of the names Kechia, Granularia, etc., is to be regarded as 

 establishing them as new genera of animals. 



Summakt. 



The tubes formed of fish-debris, found throughout the Chalk of 

 Englandj described by Mantell (1822) as Murcena (?) lewesiensis, 

 refeiTed by L. Agassiz (1844) to Dercetis elongatus, and placed among 

 Annelida Tubicola as Terebella letvesiensis by William Davies (1879), 

 are here described with more precise detail and retained in the 

 position assigned to them by Davies. A diagnosis is provided and 

 a holotype selected. 



A specimen from the Gault, of similar nature but larger, is made 

 the type of a new species, ' Terebella ' lutensis (Fig. 6). 



Similar tubes from the Cenomanian, built of Conifer debris and 

 Echinoderm debris, are described, and the question whether these 

 differences of composition indicate a difference of species is discussed, 

 but left open for the evidence of further material. Such tubes may 

 for the present be known as ' Terebella'' cf. leivesiensis (Figs. 7, 8). 



