PEDIGREE GP MAMMALS. 2/1 



and loops, between which grow simihir offshoots and 

 appendages of the blood-vessels of the ovary, so that 

 through the walls of the contiguous blood-vessels an 

 abundant exchange of fluids takes place between tlie 

 two, and therewith a prolonged nutrition and a further 

 and more complete development of the fcttus. The 

 higher character of the placental mammals, usually 

 plainly evinced by their anatomical relations, is thus 

 based on the existence of the placental mass. All inter- 

 mediate grades are, however, wanting which would 

 entitle us to infer with certainty the direct transition 

 from implacental to placental mammals. The Edentata, 

 (Bruta), manifestly the lowest of placental mammals, 

 are so devoid of any nearer morphological relations 

 with the Marsupials, that we must needs be content to 

 assume generally, on these indications, supported by 

 geographical distribution and geology, that they repre- 

 sent a very ancient branch of the placental mammals. 

 As we saw in the tenth chapter, they are scattered 

 remnants which can only by compulsion be united 

 into a single order. Sloths, armadilloes, ant-eaters, 

 differ from one another at least as much as rodents, 

 insectivora, and bats. The doctrine of Descent is not 

 discredited because it is unable to account fur these 

 fragments of bygone animal life, but in the absence 

 of data it is for the time in presence of an impos- 

 sibility. 



To ascertain the relationships of other orders, the 

 modern systematizers, and also the supporters of the 

 system of Descent, have thought fit to lay great stress 

 on the presence or absence of the so-called decidua. 

 This requires a short explanation. In many orders of 



