292 H. Woodward— On the Structtire of Trilohites. 



pendent judges. Our opinion has been submitted to Mr. Billings, 

 and by his request it is here published. The conclusion to which 

 we have come is that the organs are not legs, but the semicalcified 

 arches in the membrane of the ventral surface to which the foliaceous 

 appendages or legs were attached. Just such arches exist in the 

 ventral surface of the abdomen of the Macrura, and to them the 

 abdominal appendages are articulated. (See Plate VIII., Fig. 2). 

 This conclusion is sustained by the observation that in one part of 

 the venter three consecutive parallel arches are distinctly connected 

 by the intervening outer membrane of the venter, showing that the 

 arches were plainly in the membraiie as only a calcified portion of it, 

 and were not members moving free above it. This being the fact, 

 it seems to set at rest the question as to legs. We would add, how- 

 ever, that there is good reason for believing the supposed legs to 

 have been such arches in their continuing of nearly uniform width 

 almost or quite to the lateral margin of the animal, and in the addi- 

 tional fact that, although curving forward in their course towards 

 the margin, the successive arches are about equi-distant or parallel, 

 a regularity of position not to be looked for in free-moving legs. 

 The curve in these arches, although it implies a forward ventral 

 extension on either side of the leg-bearing segments of the body, 

 does not appear to afford any good reason for doubting the above 

 conclusion. It is probable that the two prominences on each arch 

 nearest the median line of the body which are rather marked, were 

 points of muscular attachment for the foliaceous appendages it sup- 

 ported. With the exception of these arches, the under surface of 

 the venter must have been delicately membranous, like that of the 

 abdomen of a lobster or other macruran. Unless the under surface 

 were in the main fleshy, Trilobites could not have rolled into a ball." 



In order more clearly to explain to our readers the nature of the 

 conclusion arrived at by Professors Dana, Verrill, and Smith, I 

 have re-drawn Mr. Billings's Asaphus on Plate VIIL, Fig, 1, and 

 have placed beside it a ventral aspect of the abdominal segments of 

 the "Norway lobster" {Nephrops Norvegicus, Leach). 



Let us now proceed to compare Figs. 1 and 2. Bach of the semi- 

 calcified arches of the sternum s, (Figs. 2 and 2a) is, it will be 

 observed, firmly united to the margins (epimera) of the tergum (t) 

 of the corresponding segment forming one somite or body-ring 

 (Fig. 2a). Eeferring to Fig. 1, does it not seem improbable that if 

 these so-called legs are homologous with the sternal arches in Fig. 

 2, that they should retain their normal position near the median line, 

 where, if they be sternal arches, they must have been loholly un- 

 attached, and be at the same time widely removed from the lateral 

 border of the particular segment to which each must have been 

 anchylosed ? 



Taking this fact alone, we are convinced that these appendages 

 were attached along either side of a median line, and that the lateral 

 extremities were free. 



The statement that " three consecutive parallel arches are dis- 

 tinctly connected by the intervening outer membrane of the venter, 



