64 BULLETIN 89, UNITED STATES NATIONAL MUSEUM. 



side. Those belonging to the left are retained in the matrix so nearly in their 

 relative positions that they give a fairly correct idea of the changes that occur in 

 their length, shape, and other characteristics. It appears that they represent the 

 series from the thirteenth vertebra back to the sacrum. The fourteenth rib of this 

 side is missing, but is present on the opposite side and stdl articulated (X, fig. 23). 

 The ribs of this series increase in length and in other dimensions quite rapidly 

 from the first to the fifth, when they continue subequal in length to about the 

 ninth. Posterior to the ninth they shorten rapidly, at the same time becoming 

 more slender throughout the head and shaft. Their gradation in length is well 

 shown by the measurements of the following ribs: 



mm. 



Second of series, left side, greatest length about 760 



Fifth, of series, left side, greatest length 1, 040 



Seventh of series, left side, greatest length about 1, 050 



Tenth of series, left side, greatest length about 885 



Thirteenth of series, left side, greatest length about 385 



The upper external surfaces of the anterior ribs are broad and flat, and posterior 

 to the fifth, produced posteriorly into a rather short blade, and anteriorly into a 

 somewhat shorter projection. Continuing posteriorly these projections become 

 subequal, so that a cross-section near the proximal end would be T-shaped, with 

 the cross of the. T formed by the external surface of the rib. The lower half of the 

 rib is flattened and in cross-section would be an elongated oval. Lull 1 comments 

 on the T-shaped cross-section of the ribs in Stegosaurus as follows : 



A beautiful mechanical device is shown in that the transverse process is triangular and the rib 

 T-shaped in cross section in the armor-bearing region, giving the maximum of strength, a wide bearing 

 surface, and a minimum expenditure of material. 



The significance of the above interpretation appears to be nullified somewhat 

 when it is known that almost identical conditions are found in Tracliodon, an 

 unarmored reptile. 2 Some of the Sauropoda also show a T-shaped cross section of 

 certain ribs. From this evidence, therefore, I can see no facts to bear out the 

 theory that this part of the internal skeleton was modified to support the dermal 

 armor. The inner surfaces of all the ribs are convex. The distal ends are never 

 spatulate, though sometimes thickened. Those of the median part of the series 

 are beveled off on the anterior margin. In the posterior part of the series the ends 

 are more truncate. 



The tubercle and capitular facets are placed well apart throughout the entire 

 series. The tubercle rises but little above the proximal surface, while the capitular 

 process remains fairly constant in length and is given off nearly at right angles to 

 the shaft, so that when articulated with the vertebra it forms an extremely flat, 

 lank body cavity. In this respect Stegosaurus bears a striking resemblance to 

 Diplodocus as depicted by Holland. 3 



i Amer. Journ. Sci., vol. 29, 1910, p. 209. 



2 "The oblique T-shaped cross-section that has been specially noticed in Stegosaurus, but seems, as far as I am aware, to be 

 present also in other members of the Orthopodous order. Its origin may therefore have to be explained otherwise than through 

 the weight oi the dermal armor." — F. Baron Nopcsa, in Geol. Mag., vol. 8, 1911, p. 111. 



3 Amer. Naturalist, vol. 44, 1910, fig. 17 (5). 



