PREFACE. xv 



primitive enteric .sac gives off the foundations for a variety of other 

 structures, so that what is left of it as the central element of the alimentary 

 canal is a changed and broken-up enteron, which maybe called "metenteron" 

 as opposed to the unchanged " archenteron." 



It is to these three morphological factors then, the metenteron, the 

 stomodseum, and the proctodeum, that we are called upon to assign the 

 various adaptational swellings, constrictions, and outgrowths of the alimen- 

 tary tract of higher animals. 



These distinctions arc not recognised in Professor Gegexbaur's work. 

 It will be sufficient here to point out that the exact limit of stomodieum 

 and of proctodeum in any particular case, can only be ascertained by 

 direct observation of the process of development. The metenteron is that 

 part of the alimentary canal with which the most important digestive 

 glands are connected, such as the liver, and from its walls they arc 

 formed as outgrowths. The stomodeum gives rise to salivary glands, 

 and usually to masticatory sacs (gizzards), but these latter may form also 

 in the metenteron. 



The proctodeum forms the cloacal chamber, where such exists, and 

 always receives the openings of glands (such as the Malpighian filaments of 

 insects) which are excretory rather than accessory to digestion. 



These explanations will be sufficient to make clear to the reader the 

 sense in which the words " enteron " and " enteric " have occasionally been 

 employed in the translation. 



Classification. — At the present day, naturalists have learnt to recog- 

 nise in their efforts after what was vaguely called the " natural " system 

 of classification, an unconscious attempt to construct the pedigree of the 

 animal world. The attempt has now become a conscious one. Necessarily 

 classifications which aim at exhibiting the pedigree, vary from year to year 

 with the increase in our knowledge. They also vary according to the 

 importance attached by their authors to one or another class of facts as 

 demonstrating blood-relationships. Probably no two zoologists of the 

 present day would agree, within wide limits, as to the classification which 

 comes nearest to expressing the pedigree. Accordingly it is by no means 

 desirable that students should be taught to accept any one scheme of clas- 

 sification as finite. They should be taught to look upon these schemes as 

 the condensed expression of an author's views — as the epitome of his 

 teaching, facilitating the recollection and the comparison of conflicting 

 solutions of the vast series of unsolved problems of morphology. 



I propose here, for the convenience of the student, to place side by side 

 the general outlines of the schemes of classification adopted by Professor 

 Huxley in 18G9 (No. I.), that adopted by Professor Gegenbaur in the 

 present volume (Xo. II.), and that which I have made use of in my lectures 

 during the past year (Xo. III.). 



I have taken the older classification adopted by Professor Huxley rather 

 than that more recently put forward by him, because it is one with which 



