400 ARTHROPODA. 



logical. As Johannes Miiller first iiointed out, the whole compound 

 eye forms but a single erect picture composed of separate images 

 of small area formed by the sejoarate ommatidia. This ' mosaic 

 theory' has completely rejjlaced the view that each ommatidium 

 formed a complete inverted picture. 



While the number of ocelli varies, the compound eyes are almost 

 always two in number. Where, apparently, as in the Uaphnida, 

 there is but one, there is in reality a fusion. There is also con- 

 stantly present a large optic ganglion where the optic nerve enters, 

 but lying outside the eye itself. 



The tactile organs, consisting of tactile hairs (fig. 77), are uni- 

 form in structure. On the other hand the senses of hearing, taste, 

 and smell are subserved by varying organs. It is to be regretted 

 that we know but little of these senses in arthropods, although 

 beyond question they are frequently well developed. The sense of 

 smell resides chiefly in the antennae and in the palpi of the jaws. 

 The organs are olfactory cones (modified hairs) which frequently 

 lie in pits in the skin. Similar organs in the mouth are probably 

 connected with taste. As organs of hearing (? equilibration] besides 

 the otocysts of the Podophthalmata and the tympanal organs of 

 the Orthoptera, the widely distributed ' chordotonal " nerve ends 

 of insects are to be mentioned. 



Concerning the alimentary canal it need only be said that the 

 larger proportion of it is formed of ectodermal stom odeum and 

 proctodeum, while the entodermal portion (mesenterou) forms on 

 an average but one third of the total length. At ecdysis the chitin- 

 ous lining of these parts, including the large chewing stomach, is 

 cast with the rest of the integument. The entire absence of cili- 

 ated epithelium is noteworthy. Ciliated cells have never been 

 found in arthropods. 



The most constant portion of the circulatory sj'stem is the heart, 

 which usually lies immediately beneath the back and is enclosed 

 in a more or less distinct sac which, although called pericardium, 

 is not a part of the ccelom. From the pericardium blood passes 

 into the heart by openings right and left, the ostia. Since the 

 margins of the ostia project far into the lumen of the heart and so 

 form folds functioning as valves, the heart itself may l)e divided 

 into a series of chambers, especially distinctly separated from each 

 other by the jirogressive contractioii of the wall (fig. (iij). The 

 chambers disappear when, with reduction of the body, the heart 

 shrinks to a sac. In small arthropods the heart together with 

 the whole vascular system may be lost. Since the Annelida have 



