Vol. 58.] MR. F, R. C. REED ON THE GENUS LICHAS, 69 
There are various subsidiary modifications. 
1. The Furrows.—(a) The posterior part of the first lateral 
furrows may become weak (as, for example, in Lichas levis and 
Leolichas illenoides). 
(5) Traces of the third lateral furrows may persist, as above- 
mentioned. This is not a secondary modification, but rather the 
persistence of a primitive character. 
(c) The axial furrows may become partially or completely 
obsolete (as in Ceratolichas gryps and Terataspis grandis). 
2. The Lobes.—(i) The tri-composite lobes may be incompletely 
defined on their inner side by modification (a) of the furrows; 
(ii) the regular convexity of their surface may be interrupted by 
modification (0) ; 
(iii) by modification (c) they may lose their separate existence 
by fusion with the fixed cheeks (as in Ceratolichas gryps). 
(iv) The anterior portion of the median lobe may become swollen 
into a conical protuberance (Conolichas), and the posterior part may 
become depressed and constricted (T'erataspis). 
The development of occipital lobes is usual at this stage, and 
they are always present in the groups Homolichas, Hoplolichas, 
Conolichas, and Leiolichas, as defined by M. Schmidt: of which 
the type-species are Homolichas depressus, Hoplolichas tricuspidatus, 
Conolichas equiloba, and Leiolichas illenoides. In Platymetopus 
lineatus and the other allied species (Lichas hibernicus, L. Holm, 
etc.) they are absent, and the first lateral furrows meet the occipital 
furrow at right-angles. 
These occipital lobes, as at the first stage, vary somewhat in size 
and shape. They may be relatively large and subtriangular (as in 
the transitional form Lichas furcifer), or they may be small and 
nodular (as in L. Eichwaldi and Conolichas wquiloba), or nearly 
obsolete (as in Ceratolichas gryps). 
Modifications of the neck-ring and the development on it of spines 
or processes are seen in Hoplolichas and Terataspis. Spines, often 
of great length, are found on the lobes of the glabella in these two 
groups and in Ceratolichas. The spinosity of the highly modified 
and aberrant forms, Terataspis and Ceratolichas, belonging to this 
stage, is comparable to that which exists in Arges armatus [and to 
a less extent in Craspedarges Wilcannic] of the first stage. 
Dr. Beecher * has pointed out that spinose forms must be regarded 
as derived phylogenetically from non-spinose ancestors, and that 
spinosity represents a limit to morphological and physiological 
variation. Evidence also proves that highly spinose organisms are 
the end-terms of lines of evolution, and leave no descendants. The 
extreme divergence of these members of the Lichadide from the 
normal types and their late geological appearance, just before 
the extinction of the family, support these conclusions. 
1 Amer. Journ. Sci. ser. 4, vol. vi (1898) p. 356. 
