2G0 PROCEEDINGS OF THE 



which we find the specimens. It seems to me that Dr Duncan 

 has taken the specific characters of H. Lyelli from the lower portion 

 of a stem with the hooklets broken off and slightly worn, and his 

 H. mirabiUs from the upper or more slender pai't of a stem, as seen 

 lying in the shale, with the hooklets in position. 



There is one important mistake which Dr Duncan has committed 

 in his description of H. mirabilis, to which I "Hash shortly to refer. 

 He states that the curved hook-shaped sj)ines or processes, which 

 stand out from the costse, were articulated to the tubercles upon 

 the costae, and he gives several figures to illustrate what he 

 supposes was their mode of attacliment. This view is not 

 warranted by an examination of several fine specimens in my own 

 collection and in that of Mr James Armstrong, which are embedded 

 in shale, and show the hooklets in position. 



These were not hooklets articulated upon tubercles, but small, 

 curved, spinous processes immoveably attached to the stem, either 

 upon the costse or in grooves. At their base these processes seem 

 to have been tubular, and when broken off and 'a little worn, as 

 seen in weathered specimens, they then present a deceptive 

 appearance as of a small rounded tubercle, with a pit in the 

 centre, which is caused by the liollow base of the spine. 



In nearly every specimen in which these delicate little hooklets 



are preserved in position, they are seen to be fractured close to 



their attachment with the stem. This has been produced by the 



pressure to which the corallum was subjected while lying in the 



soft shale. It is easily seen, from the irregular way in Avhich they 



are fractured, that they were broken off" by pressure, and not by 



any process of disarticulation. On some specimens there is still to 



be seen, at rare intervals, a single spine attached by its solid base 



to the stem, while on numerous other examples, where the hooklets 



have-not been fractured quite close to the stems, their bases are seen 



projecting a short distance from the costse. In one specimen the 



booklet has been broken off" near the stem, but the matrix retains a 



cast of the detached spine ; the contour of which, from base to apex, 



is perfectly even and unbroken, an appearance incompatible Avith 



the alleged mode of attachment. The above facts clearly show 



that there was no articulation of tlieir bases upon rounded 



tubercles, which would, it appears to me, be quite an anomaly in 



the structure of a zoophj^te. 



