The precise information furnished by the Limulus muscle concerning the relation of the ground membrane to the myofibrilke is of prime importance in respect to the interpretation of occasional serrated disks of apparently normal cardiac muscle and the still more complex disks of hypertrophied heart-muscle. This general type of disk is of very irregular serrated structure. The matter which was difficult of inter pretation hitherto was the nature of the cross-connections, in the form of delicate membranes, between adjacent elements of the disk proper. In view of the data now available, namely, the close connection of the myofibrilke to the telophragmata and the divisibility of the so-called myfibril units into still finer fibrillae, the matter becomes clear. The essential condition in muscular hypertrophy is an increase in size of the fibers due fundamentally to a longitudinal splitting of the myofibrilbe. Keeping in mind the connection of the myofibrils with the membrane, and the unequal tensions (relative or absolute) under which the adjacent muscle-fibrils work in the hypertrophied heart, the series of changes from the simpler comb type of disks to the complex serrated type of hypertrophied muscle become intelligible (fig. 16).
Heart-muscle, then, is clearly a syncytial structure in vertebrates and Limulus, and intercalated disks are presumably invariably present (the matter has not yet received attention below selachii), representing contraction bands which have become incapable of reversion, and thus undergo structural and chemical changes.
We are now in a position also to bring into harmony the discrepant descriptions of the relation of the simpler intercalated disks to the telophragmata. Heidenhain (12) describes the disks as invariably bounded on both sides by a ground membrane. They have also been variously described by other observers as bordered only on one side by a telophragma; and as having no definite relationship to these mem branes. My own observations on mammalian cardiac muscle led me to conclude that the disks are generally bisected by a telophragma; occasionally they appear bounded on one side by this membrane; and generally in favorable instances the disks can be seen to shade laterally into a Z-membrane. If an intercalated disk of the comb type is cor rectly interpreted in terms of a contraction band, as I have maintained, then it becomes a simple matter to explain the usual relationship of the Z-membrane to the disk, as also the several occasional variations when the membrane forms either one or both boundaries. The Z membrane forms also the mid-portion of the contraction band. As an "irreversible contraction band," an intercalated disk also is there fore usually bisected by the Z-membrane. It may conceivably occa sionally happen that only half of the contraction band may be able to relax, whereas the other half may become converted into an inter calated disk. Such a disk would then be bounded only on one side
by a ground membrane. If these same conditions should prevail in two successive contraction bands, involving the adjacent halves of the bands, then the resulting disk would be bounded on either side by a Z-membrane. Such disks would be rare, the prevailing type being bisected, or bounded on only one side, by the membrane.
The presence of a sarcolemma distinct from the endomysium is proved by the difference in staining reaction of the more delicate inner and coarser outer layers of the closely enveloping connective tissue. The outer reacts to specific stains for connective tissue; the inner does not so react, at least not to the same degree, but always to the same degree to which react also the telophragmata with which this inner layer is in close structural connection. This inner layer is a true sarcolemma. That it represents the original cell-membrane seems proved by the fact that it is connected with the nuclear wall through the telophragmata. The close association of sarcolemma and endomy sium must be emphasized; but this is exactly the same as obtains in the case of the skeletal muscle, and exactly the same structural condi tions obtain. In neither case can there be any doubt about the presence of a sarcolemma. The fact that one is present in Limu/us cardiac muscle should go far to remove further skepticism regarding its actual presence in vertebrate heart-muscle.
Neither in the cardiac nor the abdominal-spine muscle of Limulus is there any indication of a mesophragma. Thulin records the absence also of a mesophragma, and even of a telophragma, in the wing-muscles of certain insects, birds, and bats. The meaning of this structural peculiarity in these muscles is obscure. Heidenhain (22) claims that a mesophragma is present even in human heart-muscle. The occurrence of such a membrane in cardiac muscle is disputed by many histologists. When apparently lacking, Heidenhain believes that it has a thickness of less than 0.2 micron, the limit of microscopic resolution, hence indiscernible. It seems obvious that the question as to whether heart muscle anywhere actually contains mesophragmata demands reinvesti gation. If it were actually present and of the same nature as the telo phragma, and if it had the same relation to the myofibrils and to the sarcolemma, as Heidenhain believes, then it should cause festoons in the sarcolemma like those caused by the telophragma, which is not the case. Furthermore, if present, it should produce a similar folding to that produced by the telophragmata in certain distorted fibers, as shown in figure 4, which also is not the case.