245 
dary calicles in the central part leads to irregularities and excre- 
scences and then the radial arrangement of the septa is only visible 
in the marginal part. It is impossible to separate the specimens 
with radially arranged septa from those with septa with an irregular 
course, for the two extremes are connected by all possible trans¬ 
itional stages. Moreover in the small specimen of Doderleinia 
irregularis, described and figured by Gardiner (1909, PI. 39, 
fig. 28), the septa form more or less straight lines radiating from 
the central part of the corallum. The irregularities found in this 
specimen are only due to regeneration. 
A central calicle of larger size than the secondary ones is found 
in a large part of my specimens, especially in those with distinctly 
radiating septa, but these two characteristics are not always found 
together. A columella has developed in nearly all specimens at 
least in some of the calicles, it may be almost rudimentary or 
projecting distinctly in the centre of the axial fossa. The depth of 
the fossa varies in connection with the height of the septa. 
The thickness of the corallum is also subject to strong variation, 
as a rule the strongly arched specimens are much thicker than 
the flatter colonies. In the larger specimens the thickness of the 
corallum at some distance of the margin varies between 15 and 
25 mm. 
For the reasons given above I unite the forms robusta and 
irregularis into one species, in which also the form described by 
van der Horst as Doderleinia sluiteri must be included. I have 
in my material a number of specimens showing the same peculia- 
rities (small height, few secondary calicles, marginal part with long 
uninterrupted parts of septa, corallum thin) as given in the diagnosis 
of D. sluiteri. These peculiarities are found in a great many young 
specimens and in those older ones which have not assumed an 
arched form (figs. 108, 114). The marginal part of these coralla 
enlarges in a horizontal direction and is usually very irregular in 
shape, especially in very large specimens (fig. 113). In the arched 
specimens the weight of the whole colony is supported by the 
margin, in these flatter specimens the margin projects freely (the 
lower surface of the corallum touches the sea-bottom in many places) 
and may grow faster than in the arched specimens. This will be 
probably the cause of the different shape of the marginal part in 
