On Ceratopora, the type of a new family of Alcyonaria

In the introduction to the British Museum Catalogue of the Jurassic Bryozoa (1896), Gregory remarks that, “to the palæontologist, who cannot check his conclusions by the evidence of vascular anatomy or embryology, these tube-dwelling animal are a vexation and a puzzle.” This passage has reference to the difficulties that the palæontologist meets in determining the proper systematic position of many fossils that are known to us only by the tubular skeleton that they have left deposited in the rocks. Simple or colonial tubular skeletons, or more correctly shells, may be formed for the protection of recent sedentary animals belonging to the Protozoa, Cœlenterata, Annelida, Polyzoa, and Mollusca, and in many cases the only trustworthy guide to their systematic position is to be found in the study of the soft structures that formed the shell, the shell itself affording no distinctive characters. In some cases the presence of septa, and in others of tabulæ, may indicate affinities; even these characters may be misleading and give rise to erroneous conclusions. The presence of septa-now called pseudosepta-in Helipora led to the erroneous conclusion that Heliopora was a Zoantharian coral, and the presence of tabulæ in Millepora led to the classification of the Milleporidæ with other tabulate corals. It was not until Moseley examined the soft arts of Heliopora, and until Agassiz examined the soft parts of Millepora, that these corals were assigned to their proper in the animal kingdom.