The innermost layer of the ectosarc is often channelled by a system of canals,[158] usually inconspicuous, as they discharge continuously into the CONTRACTILE VACUOLE; but by inducing partial asphyxia (_e.g._ by not renewing the limited supply of air dissolved in the drop of water on the slide under the cover-glass), the action of the vacuole is slackened, and these canals may be more readily demonstrated. The vacuole, after disappearance, forms anew either by the coalescence of minute formative vacuoles, or by the enlargement of the severed end of the canal or canals. The pore of discharge to the surface is visible in several species, even in the intervals of contraction.[159] The pore is sometimes near that of the anus, but is only associated with it in Peritrichaceae, where it opens beside it into the vestibule or first part of the long pharynx, often through a rounded reservoir (Fig. 60, _r_) or elongated canal. The ENDOSARC, in most Ciliates well differentiated from the ectosarc, is very soft; though it is not in constant rotation like that of a Rhizopod, it is the seat of circulatory movements alternating with long periods of rest. Thus it is that the food-vacuoles, after describing a more or less erratic course, come to discharge their undigested products at the one point, the ANUS. {144}In a few genera (_Didinium_, for instance) the course from mouth to anus is a direct straight line, and one may almost speak of a digestive tract. In _Loxodes_ and _Trachelius_ (Fig. 56) the endosarc, as in the Flagellate _Noctiluca_ (Fig. 48, p. 133), has a central mass into which the food is taken, and which sends out lobes, which branch as they approach and join the ectoplasm. The endosarc contains excretory granules, probably calcium phosphate, droplets of oil or dissolved glycogen, proteid spherules, paraglycogen grains, etc.