new " Worm ^^ from Lower Ludloio Beds. 125 



The American hoiizons are near tlie top of the Orcloviciaii, 

 and correspond approximately to our Lower Ashgillian. 



The ori<iinal generic diagnosis is expressed in vague terms, 

 and needs interpretation with aid of the figures and descrip- 

 tions of the species. Tiie actual length observed varies from 

 1^ inch (say, 28 mm.), as in a young P. covingtonensis, to 

 6 inclies (say, 152 mm.) in an example of P. tenuis. Tho 

 actual width observed in the compressed fossils varies from 

 " one fourth of a line" (say, 0*5 mm.), in the smallest P. tenuis, 

 to about 2 mm., as seen in the figure of P. simplex. 



Owing to the incompleteness of most of the specimens, the 

 ratio of width to lengtii cannot be calculated witii certainty. 

 It is, however, possible to calculate tlie relative height of the 

 segments, on the basis of such measurements as are provided, 

 though these are not very precise : — 



Absolute height Eatio to 



of segment. width. 



P. tnagnus 0-12 10/100 



P. ornafHs 0-23 12/100 



P. covingtonensis 015 15/100 



P. siynplex 0-5 33/100 



P. tenuis 0-5 50/100 



The nature of the segmentation is not clear. Ulrich's 

 figure of P. simplex (op. c'lt. pi. iv. fig. 1) probably represents 

 the "complete individual" mentioned on p. 91. This has a 

 length of about 19 mm., a greatest width of 2'7 mm., and 

 tapers rapidly at each end. The drawing shows thirty-two 

 segments, and, since the specimen is bent round so that one 

 end almost approaches the other, these segments are lower 

 on the inner side of the curve than on its outer, and the 

 draughtsman has represented them as imbricating. This 

 important feature is not alluded to in the text, nor is it 

 suggested or mentioned under any other species. It would, 

 of course, be particularly obvious in a form with the high 

 and well-marked segments of P. simplex. 



The segments are ['.apillate in P. ornatus and P. mognus ; 

 in all other species, including the genotype, they are described 

 as smooth. In P. ornatus the papillfo form either one row 

 in the median line of a segment or one row near each border 

 ofasefMuent. Ulrich's en Jargred fioure 16 shoAvs about twelve 

 papillae in each row, all closely set ; that means about twenty- 

 five in the conijjlete circle of each segment. In P. magniis 

 " each segment is ornamented with a single row of six or eight 

 papillse " (j. «., twelve to sixteen in the complete circU'). 



We pass now to the first record of the genus from this side 

 of the Atlantic, and the first occurrence outside the Ordovician. 



