‘ { ‘ oan pace are rai Vad baw veer ete pewa pre ee) oe tenes oon on re ' ran Terie eer Ptah an i toe a , e os ice te eo men . Rie eh aque cael ‘ an are . oe te ‘ ae Oe 8 HE | an . ‘ ‘ . Darwen 4 ee ee poi oe , . . snk ’ 1. aie pee free ee ewe ete ‘ Vier gurney pee tes ‘ ree ie Cp ee ee te ‘ ‘ apie ‘ wees . F era aD 8 oe oe oe fa weve Gh Bois wee eed fet tt ew ee ae . rarer eta rare) . poe ‘ ere tee) ne ' ted awe ee owe ns sheen a a omgte se ‘ : . feast se ou eaten “4 eee 4. Copper as Cupric Xanthate—Cu(C,H,S,O).. Redissolved ash from A, in the presence of potassium ethyl xanthate should give a yellow color due to the formation of copper xanthate 84 OTTO GLASER. The test which is sensitive roughly to’ 1 part in 50 million, is not vitiated by the presence of sea-salts nor by small amounts of iron, lead, nickel, cobalt, zinc, or manganese. Results: Positive and very decided. 5. Copper as Crystalline Cupric Sulphate—CuSO,-5H,O. After precipitation of the iron as hydroxide the ash was redissolved in sulphuric acid and drops of the solution allowed to evaporate to dryness on a slide. Under the microscope long blue needles corre- sponding to copper sulphate. The blue color greatly intensified by ammonia. 6. Copper as Metallic Copper—Cu. 5 cc. of A plus excess NH,OH. Copper if present in sufficient quantity should be deposited on tin-foil. Result: positive but faint. With aluminum, however, despite incompleteness, the deposition was very marked; film, rose-colored by reflected light; bright and typically copper-colored when polished. 7. Copper by Electrolytic Deposition. 10 c.c. of A. Platinum Electrodes. Voltage 2.6; amperage 1/10. At this tension copper and only copper can be deposited on the kathode. After 24 hours, result positive and distinct. Could not be weighed. Redissclved deposit and got positive test with.potassium ethyl xanthate. On the basis of this evidence it was no longer possible to doubt the existence of copper in the normal Arbacia egg. The basic necessity for a secretion of copper compounds or copper-bearing substances is therefore given. If such materials are held up tem- porarily by the chorion, the differences between normal and de- chorionized ova, noted in sections IV. and V., would be explained. But do the eggs really secrete copper in any form? We could perhaps answer this question if we could associate the metal with definite structures and substances in the egg. VII. LocaALizATION OF CoPpPER IN THE Arbacia Ecc. The problem calls for the localization of the copper. I em- ployed direct methods as well as indirect. The eggs used were shed under the conditions previously mentioned and fixed in abso- lute alcohol. J studied both whole mounts and sections, 5m in thickness. The latter, after removal of the paraffin, like the un- sectioned eggs, were treated with various reagents and examined in the opaque condition, wet or dry; or after clearing, as trans- parent objects, in glycerine or balsam. All the reagents were tested and again found negative for copper. COPPER, ENZYMES, AND FERTILIZATION. 85 1. Copper in the Egg Pigment. Since copper is an integral part of so many organic coloring matters, it was natural to examine first of all the pigment of the egg. This material, as is well known, is widely distributed in granular form in the cytoplasm. As the granules are most nu- merous in tangential sections, the pigment bodies are evidently concentrated near the surface. This also seems to be true, to some extent, immediately about the nucleus. It is entirely justifiable to identify these granules as the source of the coloring matter secreted by the eggs. I consequently pre- cipitated the pigment from normal egg secretion by the chloroform method described in an earlier paper (21°). After incineration an alkaline solution of the ash in twenty minutes gave a relatively heavy, though incomplete, deposition of metallic copper on alumi- num. The deposit was dissolved and gave a deep yellow color on addition of potassium ethyl xanthate. Direct proof of copper in the pigment bodies sn situ is also pos- sible. Tangential sections which contain more of these granules than others are the very ones that stain most heavily with hema- toxylin. This we should expect in the presence of copper, though the test’ is not dependable if iron is also present. I therefore adopted a procedure employed in the microchemical analysis of minerals. The method depends on the formation of a triple nitrite of potassium, copper, and lead (’94). To sections on the slide I added a trace of sodium acetate and a somewhat larger quantity of potassium nitrite. The whole was then acidulated with acetic acid. Last of all, a few grains of lead acetate. The presence of copper, as little as 0.05 microgram, if crystals are desired, completes the conditions necessary for the formation of the triple salt, K,CuPb(NO,),, which is jet black. In this test, which differentiates between iron and copper, the sections most darkened are the very kind most affected by haema- toxylin. There is a marked graying of the cell contents to which a partial reduction of the pigment bodies contributes. The gen- eral effect is due to a decided increase in cytoplasmic granulation— the granules being excessively minute and, like the microsomes of 11 microgram (ugr.) = 1/roocoth milligram. 86 OTTO GLASER. the cytologists, black. If the material is kept moderately warm on a water bath and the reaction is permitted to continue for half an hour, the pigment bodies themselves become distinctly bluish or even black. This suggests a gradual unmasking of the copper in Macallum’s sense (12). 2. Copper im the Egg Membrane. A definite visible membrane (’131) invests the unfertilized Arbacia egg and constitutes a barrier through which pigment must pass in order to reach the outside. If now the pigment itself con- tains copper, the same thing might easily be true of the membrane through which it passes. | But the membrane is by nature yellowish-brown. This excludes tests depending on the formation of cupric cyanide, ferrocyanide, or xanthate. With hematoxylin a bluish tint develops, and this could be interpreted as evidence of copper if iron were absent. However, since the latter is almost certainly present, we must fall back on the triple-nitrite reaction. In this the original native yellow-brown of the membrane is replaced with black. Under low powers the membranes are in very sharp relief. The oil immersion, however, resolves their uniform black, in optical section, into a series of irregularly spaced discontinuous beads connected by an exceedingly thin continuous black line. This test was repeated on several sets of sections. The results were uniform and without inconsistencies. The evidence, then, that copper exists in the Arbacia egg membrane appears valid; yet the quantities involved are so small that I felt impelled to check the results on the hen’s egg, whose vitelline membrane, I reasoned, should contain copper if the analysis of the yolk by Fleurent and Levi (loc. cit.) is correct. This supposition was readily substantiated by four different methods. The vitelline membranes were removed from the yolk and washed in a stream of running tap-water for twelve hours. In this way all but negligible quantities of yolk were removed. After incineration a solution of the ash, upon the addition of ammonia, turned blue. This test was supported in neutral solu- ee ee ee. COPPER, ENZYMES, AND FERTILIZATION. 87 tion by a marked yellow brought about by potassium ethyl xanthate. Again the triple-nitrite test was positive. Under the microscope the treated membranes showed numerous irregular blotches and specks of black. By prolonging this treatment for thirty minutes and following it with a short exposure to 100° C., the membranes became closely speckled with black and underwent a discoloration quantitatively more in harmony with the very marked reaction gotten with ammonia and potassium ethyl xanthate. The copper in this membrane is, therefore, also “‘ masked”; moreover, the dis- covery of copper in the Arbacia egg membrane does not stand alone. 3. Copper sn the Chorion. Since the pigment during its outward passage permeates the chorion, the latter necessarily contains traces of copper. There is also some direct evidence of this, for hematoxylin imparts to the jelly a light bluish tint, whereas the triple-nitrite test brings out a few scattered granules stained black. Infinitesimal quantities of copper, then, can be demonstrated directly in the chorion. 4. Copper and the Cortex. Inasmuch as Lillie (Joc. cit.) has assumed the presence of a copper-avid substance in the cortex of the egg, I paid especial attention to this region both in untreated sections and in those exposed to reagents for copper. Unquestionably the cortex differs from the remainder of the egg. Yolk, pigment, and other visible granules are absent. The zone immediately beneath the egg membrane, even at very high magnifications, seems to be optically clear. In sections not ex- posed to reagents for copper the cortical layer appears blue—no doubt the result chiefly of refraction. Nevertheless the color is intensified by ammonia and hematoxylin and does not disappear entirely when viewed by light transmitted through ground glass nor after the sections have been cleared in glycerine, xylol, or balsam.? Furthermore, perhaps on account of distortions through shrink- age, the blue layer, instead of being in immediate contact with the _2There are, of course, other effects of refraction throughout the egg. 88 OTTO GLASER. membrane, not infrequently lies at a considerable distance beneath it. All this is indicative of a region differentiated from the re- mainder of the egg. | Is copper demonstrable in this layer? None of the direct methods could give an unequivocal answer, though the triple- nitrite test did reveal a few black granules similar to the black beads found in the membrane. This might be considered a dem- onstration of copper if the displacement of an occasional pigment granule in the process of sectioning could be altogether set aside. But there are other considerations. The fertilization of egg fragments has been reported by numer- ous investigators, including myself (’13+), yet the correctness of these observations has been doubted because some egg fragments are incapable of fertilization. The difference is accounted for in the work of Chambers,? who finds that fragments derived exclu-_ sively from the interior of the egg can not be fertilized, whereas those containing a fair portion of the cortex respond like the origi- nal egg. This evidence, if not compelling, at least indicates more immediately than any other the physiological distinctiveness of the cortex and the necessity of cortical materials in fertilization. ' Now, certain secretions of the egg are likewise necessary. This was first shown by F. R. Lillie (’13?), and subsequently by my sterilizations with charcoal (’21*). A logical combination of our two necessities traces the secretions to the cortex. But are they— or perhaps better, their forerunners—really there? For the nor- — mal egg this has not been demonstrated, nor is the cortex, in all probability, their only location; for egg fragments, with chorion and membrane both removed, the inference appears less hazardous. The concentration, then, of the secretions, or their forerunners, in the cortex of the egg may be assumed as a not unreasonable work- ing hypothesis. On the basis of isolable precipitates, Miss Woodward (’18) and I (’21°) have contended that these secretions contain at least two separate substances—a lipolytic ferment and a material which ag- glutinates spermatozoa. The immediate problem consequently narrows down to this: Is copper demonstrable in precipitated lipolysin and agglutinin ? 3See Lillie’s ‘‘ Problems of Fertilization,” p. 264. COPPER, ENZYMES, AND FERTILIZATION. 89 A. Lipolysin—l, The lipolysin first tested was prepared as follows: the eggs were permitted to secrete as usual; the agglutinin was salted out with ammonium sulphate, which incidentally also precipitates much of the pigment. After filtration, the lipolysin iteslf was thrown down by means of barium, redissolved in dilute hydrochloric acid and precipi- tated by acetone which unless in great excess holds the remaining agglutinin in solution. The lipolysin so secured was a white powder free from pigment, though in the course of time it turned’ slightly purple. ~ After incineration a solution of the ash gave a positive reaction for copper with potassium ethyl xanthate. Lipolysin—II. This precipitate had exactly the same history as the foregoing except that charcoal was used in place of barium. Moreover, the powder did not change color but remained snow white and therefore may have been a somewhat purer product than lot I. _By the xanthate test, lot II. also contained copper. B. Agglutinin—l. The agglutinin first examined came from Asterias egg-secretion which had been freed from lipolysin by the barium method. The xanthate test, carried out as before, gave a perceptible reaction for copper but as the separation from the other constituents of the secretion was in- complete, this result is inconclusive. Agglutinin—ll. A second test was made, this time with Arbacia agglutinin which had been merely salted out with ammonium sulphate and contained there- fore considerable quantities of pigment. In this case the xanthate test was markedly positive. Agglutmn—ITI. Arbacia agglutinin whose separation from pigment and other im- purities was accomplished by chloroform, charcoal, and final differen- tial precipitation from acetone. This material was negative for copper. In all I examined five different lots of agglutinin, and in every case preparations whose purity I supposed to be high were the ones that proved to be either completely negative for copper, doubtful, or at most very slightly positive. It appears that copper is not an essential constituent or even regular associate of pure agglutinin. These results, together with the triple-nitrite test and the con- siderations that suggest a cortical concentration of the secretions, justify my assumption that copper is present in the cortex of the egg immediately beneath the vitelline membrane. This copper, 90 OTTO GLASER. however, is not associated with the egg pigment or the aggluti- nating material, but with the lipolysin, whose concentration in the cortex is thus one step nearer to being a demonstrated fact. VIII. Copper AND ENZYMES. Lipolysin is or contains a lipolytic ferment whose presence ac- celerates the hydrolysis of higher and lower fats’ (22?) and the synthesis of ethyl-butyrate (22%). The association of copper with this enzyme, though contrary to expectation, can not be evaded. Does the case stand alone? Apparently not. For example, S. Yagi (’10?) finds a thirty-four fold increase in the copper content of the rabbit’s liver when the animal is fed 4.8 gr. copper sulphate in assimilable doses during a period of seventeen days. This agrees perfectly with the results of Titze and Wedemann (’11) on the goat and those of Rose and Bodansky (’201), who find the copper content of the oyster’s hepato-pancreas to be twice that of the muscle. Although as yet no one has risked the suggestion, it seems apparent that copper concentrates in tissues rich in lipases and certain other enzymes. May we risk another step and say the copper is concentrated in the enzymes? Be this as it may, ‘it was these considerations that led me to test the following commercial preparations: A. Pancreatin (Merck’s). This preparation contained all the pancreatic enzymes. The ash from about .2 gr. gave a marked reaction for copper as copper xanthate. B. Pancreatin from Pig (Squibb’s). Result exactly as in A. C. Pancreatin (Parke-Davis). Result as in B and A. Is the copper associated with the lipases exclusively? This can not be determined from the above tests. I therefore tried a proteolytic enzyme likely to be free from lipases, amylases, and other enzymes. D. Pepsin (origin unknown). The ash from .2 gr. when freed from iron gave no positive indication for copper. However, the iron hydroxide was found to hold the cop- per back. I therefore washed the precipitated hydroxide with ammo- —— COPPER, ENZYMES, AND FERTILIZATION. OI site: whereupon the xanthate gave a very faint though unmistakably positive test. E. Pepsin (Parke-Davis). Also positive; faint after precipitation of copper by iron hydroxide. These tests prove that copper occurs in preparations of pancre- atic enzymes and pepsin. The association with lipolysin, there- fore, is not a necessarily isolated instance, but very possibly a special illustration of a general rule. Indeed, the studies of von Euler and Svanberg (’20?) definitely foreshadow a natural relation between enzymes and heavy metals. IX. LocarizATIoN oF Copper IN Eccs ExPoseD To COPPER SOLUTIONS. Normal eggs, in addition to secreting copper compounds, remove copper sulphate from sea-water when this salt is present. In eggs so exposed we should expect differences, if not in distribution, then at least in the concentration of the copper which they contain. Such analyses seem doubly worth while, for they seem not only to control and correct the observations on normal material, but may ultimately throw some light on the machinery of copper fixation and metabolism in general. Tests consequently were made on material with a total volume of .5 c.c. exposed for 20 minutes to 5 c.c. of a 1/180 m Cu solution in sea-water. In every other respect the eggs were treated as described in section VII. Moreover, the material was from the same lot as that used for the localization of copper in normal eggs. 1. Copper Absorbed by the Chorion. From the experiments of section IV. we expect for the jelly of exposed eggs a copper content greater than that found in the nor- mal chorion. This expectation, of course, is realized, “as can be shown by means of the hematoxylin test with entire eggs. Super- ficially such eggs take on an unmistakable blue color which micro- scopic examination localizes distinctly in the jelly. The same effect is gotten with the triple-nitrite test, only the color is then black instead of blue. 92 OTTO GLASER. 2. Copper Absorbed by the Egg Membrane. With hematoxylin the membrane of the exposed egg is bluer than the normal. The triple-nitrite test, however, is more con- vincing. In optical sections of the surface one no longer finds individual black beads strung on a thin thread, but a heavy con- tinuous black line. Even the highest powers fail to resolve the continuity. It seems very clear, indeed, that potassium-copper- lead nitrite is more abundant than in normal membranes. On this basis it is possible to understand one of Lillie’s results. If an excess of copper in the membrane alters it so that sperma- tozoa are not able to penetrate as easily as under normal conditions, we should expect “a certain virtue in mass action in the presence of this inhibitor of fertilization” (Joc. cit., p. 129). I noted ex- actly the same virtue (’I5) in connection with eggs which had been exposed to calcium. This particular point and its meaning have both been overlooked in subsequent discussions. 3. Copper Absorbed by the Cytoplasm. Inside the exposed egg there are differences in the same sense. The number of black granules after the nitrite treatment is greatly increased. As a whole and in sections the exposed egg is dis- tinctly darker than the normal. Copper evidently passes inward through the membrane and diffuses generally through the cyto- plasm. 4. Copper Absorbed by the Cortex. cé If, as Lillie’s experiments indicate, the cortex contains “un- saturated ” copper-avid material, one may expect very considerable differences in the cortical zone. As a matter of fact, the triple- nitrite test brought to light black beads similar to those seen in and immediately under the normal membrane; they were decidedly more numerous than in normal eggs and gained in distinctness by heating. Furthermore, the blue zone which in normal eggs resists resolution by all the methods employed now differentiates slightly with both potassium ethyl xanthate and potassium ferrocyanide. With the first, the blue acquires a greenish tint and becomes sep- arated from the egg membrane by an exceedingly narrow yet dis- tinguishable yellowish layer; with the second, a very thin brown band intervenes between the original blue and the membrane. COPPER, ENZYMES, AND FERTILIZATION. 93 These changes become intelligible if we assume, in the one case, the formation of minute traces of copper xanthate; in the second, of copper ferrocyanide. Since the same reagents produce no noticeable changes in these locations in untreated material, it appears that the cortex of an egg exposed to copper contains more than the normal quantity of the-metal. Moreover, the excess is localized chiefly near the surface, and since it is not the agglutinin that is copper bearing, but the lipolysin, it appears probable that the latter is also concentrated immediately under the vitelline membrane. X. Copper Map oF THE NorMAt Ecc. From all these tests—those on the controls as well as those on material exposed to copper sulphate, I have constructed a chart which indicates the distribution of copper in the normal Arbacia egg. This map is a visual summary of the chief results and infer- ences, and in view of the preceding discussion seems to require no other comment than that given in the legend under the figure. Diagram showing distribution of copper in the normal egg of Arbacia punctulata. The central clear area is the nucleus. Immediately about this, and extending to the cortex, the larger black spots represent pigment granules in which copper was demonstrated indirectly by analysis of secreted pigment and directly, in situ, as potassium-copper-lead nitrite. The latter, as well 94. OTTO GLASER. as hematoxylin, indicates also the presence of diffuse copper in the cytoplasm. This is shown by the finer stipples. The cortical layer is shown as differentiated into two zones: an inner, free from copper; an outer, immediately under the egg membrane with consider- able concentration of copper. The evidence is again the triple-nitrite test on normal eggs, and the xanthate and ferrocyanide tests on eggs which had been exposed to solutions of copper in sea-water. _ The vitelline membrane is shown as a line whose irregularly spaced black beads represent regions in which the triple nitrite was concentrated. The “ membranes of eggs exposed to copper solutions appear as continuous heavy black lines without beads. The region outside the vitelline membrane represents the chorion with its diffuse and highly rarefied copper. The diagram is intended to indicate distribution. It cannot indicate quan- titative differences accurately because a variety of tests was used and the extent of discoloration depends not primarily on the quantities of copper but on the molecular size of the copper compounds formed. These are necessarily different in the several tests. XI. Orper oF MAGNITUDE OF THE QUANTITIES OF COPPER INVOLVED. Quantitative data on the copper content can not be given. The methods for determining the copper are, indeed, reliable enough; but, unfortunately, the chorion alone prevents anyone from know- ing how many eggs are present in I c.c., and until this number is definitely ascertained our initial measurements can hardly articu- late closely with the results of chemical methods unusual in deli- cacy. For the present, then, we can arrive only at approximations. 1. The Ovarian Egg. Eleven c.c. of ovary were dissolved in 11 c.c. of concentrated nitric acid and after complete destruction diluted to 100 c.c. with distilled water. Ten c.c. of this solution were evaporated to dry- ness and the incinerated ash dissolved. After neutralization with ammonia the volume was restored to 9 c.c. To this was added 1 c.c. of a solution of potassium ethyl xanthate whose concentra- tion was I gr. per 1,000 c.c. of distilled water. This same amount was added to 9 c.c. of a copper sulphate solution containing 10 p Bin COPpPer per cc: Colorimetric comparisons were then made, and the concentra- tion, C,, of the unknown calculated from the concentration, C,, of the standard and the depth of the unknown, D., necessary to match COPPER, ENZYMES, AND FERTILIZATION. 95 a given depth, D,, of the standard. Accordingly, EGR SD: Od A Ds . Test (Cr 3 IDh Micrograms Cu. D2 per c.c. Eggs. ienp cuote meen 10 X .50 ew 20.54 2,21 lla eaten Ree 5 IO X .30 mona 1.38 IE Se tere ae 110) < 2 14.29 1.40 NUL saath sce 10) << Z 15.27 Teo Ae eaves 1@ X< TR pease 13-39 Tene Average = 16.65 mgr. 2. The Shed Egg. Ten c.c. of the material referred to as A in section VI. were treated and tested in exactly the same way. Test Gi X Di Micrograms Cu. D2 ; per c.c. Eggs. Muay sesleaiehor ats Io I x 161.25 1.55 Ie eal Oe SE nO S< wag 170.50 DB Asst eyeiras os MOVOerT 178.50 1.40 ATT eas ae cee u@ >< ar aes ars Tesi Average = 174.56 mer. 3. The Fertilzed Egg. Forty c.c. from a solution of 1.2 c.c. of fertilized eggs gave the following results : 96 OTTO GLASER. Test Cora Micrograms Cu. D2 per c.c. Eggs. Ae os eet 2 UG) 3<, X «5 20.83 2.00 Neue hale ieteeoss = 10 X .5 21.25 1.96 aA 25873 10 X .5 20.50 2.03 Average = 20.86 wer. The copper content of the fertilized ovum apparently has the general order of magnitude characteristic of the unshed ovarian egg. That of the shed egg is, roughly, from eight to ten times greater. XII. Discussion. 1. The Copper Problem. Unless bound up in respiratory or other pigments, copper is con- sidered essentially a poison. Its use in germicides and fungicidal solutions, the harmful influences on higher and lower plants, power to block organic catalysis, the medicinal properties of colloidal suspensions and copper salves—all support the prevalent view. Yet: there are no poisons in nature; there are only poisonous effects. These may be exercised by the commonest articles of diet at certain concentrations and by copper, it happens, at very low ones. Even a cursory examination of the literature suggests further misgivings. Copper has been found repeatedly in three of the five main divisions of the plant kingdom and among animals, with equal frequency, in nine phyla, ranging from protozoa—if Volvox (21°) is an animal—to man. However, only Maquenne and Demoussy (’20*), who worked on plants, and Rose and Bodansky (7201), whose studies cover a wide range of marine animals, are actually bold enough to suggest that copper in general may be more than an adventitious element which living things somehow tolerate. Specific processes in which it normally plays a role are still unknown; moreover, they are quite likely to remain so unless we free ourselves for investigation by an admission of ignorance. As a matter of fact, copper has many of the qualifications of a COPPER, ENZYMES, AND FERTILIZATION. 97 biological element. It is widely diffused in the environment and has been so since earliest geological time. Though listed among the less active metals, it is capable of entering into a large variety of combinations, including numerous organic unions. It is diff- cult to see how a living thing could avoid copper except by some definite mechanism of exclusion. The absence of this is significant, but does not establish physio- logical importance. Very possibly living things are mere sieves that hold the copper back; very possibly, too, its marked concen- tration in such tissues as the liver is a liability, by accident de- pendent on other attributes and under ordinary circumstances negli- gible. Still one would like to be certain, and in our present state of knowledge this is impossible. Yet a negative answer, excluding copper from the realm of physiological processes, even now ap- pears unlikely. It is impossible to conceive the synthesis of respi-. ratory pigments, turacin, or any other product without thinking of a long linkage of reactions inevitably affected, directly or other- wise, by their final end result. ' 2. Copper and Enzymes. Von Euler and Svanberg (Joc. cit.) have shown that the addition of saccharase results in a marked reduction of the free silver ions in dilute solutions of silver nitrate. Since the silver did not be- come colloidal under the influence of reductions, possibly due to certain constituents of the enzyme preparation, these writers sug- gest a binding of the silver ions to certain constituents of the saccharase solution. Since such unions render the enzyme inactive, why not assume that the constituent of the solution with which the metallic ions combine is the enzyme itself, an organic co-enzyme, or both? Such an assumption seems all the more reasonable because pro- longed dialysis, I find, does not remove all the copper. Thus if copper incapacitates at all, catalysis would be excluded in any case.+* 4Tt is not possible that certain differences of proportionality between the effects of silver and mercury salts could be explained on this basis? Perhaps too the recovery known as “ The Danysz Phenomenon” results from a re- distribution of metallic ions between enzyme and co-enzyme. Even if co- enzyme and enzyme together make up, and are identical with, the entity, enzyme, the allocation of Cu to different positions in this system might have results essentially those suggested both in sense and in degree. 98 OTTO GLASER. This assumption involves on the part of enzymes or organic co- enzymes a capacity for combining with heavy metals. Since this power is not only very great, to judge by the dilutions that in- capacitate, but likewise inseparable from normal ferments, we should not be astonished if enzymes from copper-bearing organ- isms or tissues give positive results when tested for copper. 3. The Oligodynamic Effects of Copper. The poisonous effects of extraordinarily high dilutions of metal- lic copper were first noticed by Nageli (’93), who found one part in seventy-seven millions rapidly fatal to Spirogyra. Because of the extremely small quantities involved Nageli spoke of “ oligo- dynamic action.” As these results have been repeatedly verified on other forms, it may be convenient to retain Nageli’s term. However, “oligo- dynamic” need imply nothing more mystical than a chemical or perhaps physical relation between extremely small quantities of material. As yet no acceptable explanation has been offered.° But there have been suggestions. Nageli was able to destroy the toxicity of oligodynamic solutions by means of paper, wool, paraffin, gums, and gelatin. The experience is quite comparable to the protective action of egg-water, gum arabic, and gelatin de- scribed by Lillie (’217). Inasmuch as the copper in the first case at least is probably present as electropositive colloidal hydroxide or carbonate, Bayliss (’15?) imagines adsorption on the electro- negative surfaces of Nageli’s detoxicants as the basis of their anti- toxic properties. Conversely, the toxic effects become explicable as the outcome of adsorption by electronegative surfaces or parti- cles critically involved in or on the affected tissues. At this point Bayliss leaves the problem. We still do not know why such minute quantities are effective. Do they cause precipi- tation? If so, how much precipitation can they conceivably bring about? And, finally, what is precipitated? The neutralization of electrical charges in all likelihood is unavoidably associated with 5 Quite recently, Saxl, as quoted by von Euler and Svanberg, assumed “ dass die Metalle eine eigenartige Keimtotende Kraft besitzen, welche nicht mit der Loéslichkeit der Metalle zusammen hangen soll.” Saxl refers to a physical energy “die sich zunachst auf der Oberflache der Metalle abspielt, jedoch auch in andere Medien tibergehen kann” (’202 p. 378). oe Pee i ee COPPER, ENZYMES, AND FERTILIZATION. 99 the ultimate mechanism of oligodynamic activity. For the pres- ent, however, our concern must be with facts a trifle more imme- diate. From these, it seems to me, we can derive a second sug- gestion. The catalytic power of colloidal platinum is destroyed by traces of hydrocyanic acid and can be restored by simple aération. Moore (’21") compares this experience of Bredig and his pupils with the effects of hydrocyanic acid on the peroxidases of the enzymic oxidation system, “If these peroxidases are responsible in tissue cells for the uptake of oxygen by the protoplasm, it may well be that the poisonous action of hydrocyanic acid in such min- ute doses is due to interference with the action of the peroxidases ” @ocsctt; p: 221). 7 Disregarding the specific case which Moore had in mind, there is implicit here a definite relationship between enzymes and oligo- dynamic action. This relation, however, is also implicit in the copper inhibitions of Lillie. Again the union of saccharase with silver or copper implied by von Euler and Svanberg, together with the natural occurrence of copper in preparations of lipolysin, pancreatic enzymes, and pepsin, now reported, I believe, for the first time, are just what one might expect if the oligodynamic properties of heavy metals are traceable to the activation or inacti- vation of enzymes. These effects must depend on the capacity of enzymes to bind the metal. Until stoichiometrical determinations have .been made nothing further can be said about such unions. But no matter what their nature may be, their occurrence should result in the presence of heavy metals in enzyme preparations when- ever such presence is a physical possibility. But the presence of copper raises an apparent difficulty. One might arrive at the paradox that enzymes, by the very nature of the case, must be normally incapacitated. This absurdity is easily dispelled. There is no reason for considering the normal enzyme as saturated with copper. For that matter, the enzyme proper may not be involved at all, for the observed effects can all be explained equally well if the copper is held by organic co-enzymes where these are present. But in this case also we are not com- pelled to consider the co-enzymes as saturated. IO0O OTTO GLASER. On this basis it is possible to explain the most puzzling feature of oligodynamic action. It is not that copper, silver, gold, or any other metals have this or that effect, but that infinitesimal amounts give results apparently out of all proportion to the quantities involved. In the case of copper, for example, I imagine the oligodynamic effect as due to the inactivation of that fraction of the enzyme or co-enzyme which was not inactivated by the copper present in the first place. Where the oligodynamic effect is produced by silver or some other metal, the total of inactivated enzyme or co-enzyme would be composed of two fractions—the one inactivated by cop- per, the other by silver. This view of the case has stoichiometrical implications which further work may or may not justify. But however this may turn out, the presence and effect of copper in preparations of normal enzymes calls urgently for further study. Quite apart from the biochemical questions that suggest them- selves at once, the discovery carries with it problems of wide bio- logical significance. Is the liability to inactivation compensated by the production of larger ‘quantities of the enzymes? Or does compensation come about by variety and differential susceptibili- ties? Why the tremendous number of enzymes when, aside from a few cases of molecular rearrangement, the only radical processes we have to deal with are the reversible oxidations and hydrolyses? 4. Copper and Fertilization. Probably most of the copper present in the Arbacia egg is incor- porated in the pigment whose elimination discolors the sea-water. As I have shown (’14), the rate of pigment secretion is increased 50 per cent. while the eggs are undergoing fertilization. This explains why the copper content of fertilized eggs is so much lower than that of unfertilized. My values for the copper content indicate in fertilized eggs an order of magnitude quite different from that in ripe eggs and essentially the same as that of the. immature ovarian ovum. ‘Tentatively, therefore, one may hazard that, among other things, fertilization restores the copper content to an order of magnitude characteristic of the unripe egg. In the absence of more accurate data it is premature to discuss enh Aa a COPPER, ENZYMES, AND FERTILIZATION. IOI this fact at length. Nevertheless we may possibly find along these lines some help in clearing up the uncertainties that now beset us. Very possibly the concentration of copper normally has something to do with checking the growth of the egg, whereas the heightened rate of secretion during fertilization restores conditions essential for further growth and development. No doubt the linkage can be pictured in several ways, but one way is this: the elimination of pigment from the egg might result in the production of fresh pigment, or some other product, which, if copper-avid, might also draw upon the copper of the enzymes and thus assist in the process of activation. All this, however, is merely in the realm of possibilities. At the rate at which discoveries are being made in this field hardly anyone would wish to formulate a theory of fertilization. The process is far more complicated than it seemed ten years ago, and if one thing is more certain than another, it is that the major classes of evidence have not yet been handed in. XIII. Summary. 1. Nearly 37 per cent. of the copper sulphate added to sea-water is precipitated at once when the concentration is Cu==/460. 2. Two tenths c.c. normal Arbacia eggs in 75 minutes reduce the concentration of 14.8 c.c. of a Cu Soul Jalen from n/1,460 to n/1,790.9. 3. In this reduction the egg jelly or chorion is heavily involved. 4. It was impossible to determine the quantities of copper ab- sorbed by normal eggs and eggs without jelly because ot an appar- ent secretion of copper by the eggs themselves. 5. Therefore, the demonstration of copper in Arbacia eggs was undertaken. The copper was identified as cupric hydroxide, cupric cyanide, cupric ferrocyanide, copper xanthate, crystalline cupric sulphate, as metallic copper on tin-foil and aluminum, and finally by electrolytic precipitation under conditions under which copper and only copper could be deposited. 6. The copper was localized in the egg directly by means of hematoxylin and the triple-nitrite of potassium-copper and lead. Indirectly it was localized by the analysis of egg secretions. 7. The copper occurs chiefly in the egg pigment; the vitelline 102 OTTO GLASER. membrane and the chorion also contain copper. It is not a regular constituent of agglutinin precipitates, but was found constantly in precipitates of lipolysin. 8. This association with lipolysin makes possible the localization of both copper and ferment in the cortex of the egg. 9. The association of copper with lipolysin is not an isolated case. Copper was found also in preparations of pancreatic en- zymes and pepsin. 10. In eggs exposed to a n/180 Cu solution for twenty minutes the copper is widely diffused through the cytoplasm and concen- trated in the chorion, the vitelline membrane, and the cortex. t1. Approximately the amounts of copper normally present in 1 c.c. of Arbacia eggs are as follows: Unripe ovarian eggs—= 17 mer. Ripe shed eggs = 175 wer. Fertilized eggs ==) it See 12. From the preceding and other considerations it is suggested that copper, in general, may be more than an adventitious element, physiological only in pigments, and merely tolerated in all other connections. 13. The association of copper with enzymes is explained as the outcome of some sort of union, very likely chemical, between en- zymes or co-enzymes, or both, and the metal. It is also suggested that possibly the differences in the proportionality between silver and mercury effects, as well as the Danysz recovery after “ poison- ing,” may be due to the distribution and subsequent redistribution of the metallic ions between enzymes and co-enzymes. 14. The oligodynamic action of copper is explained as due to the inactivation of that fraction of the enzyme or co-enzyme which was not normally inactivated by the copper present in the first place. If inactivation is produced by silver, it is suggested that the total inactive enzyme or co-enzyme would be composed of two fractions—the one inactive because of the normal copper content, the other because of the silver added. 15. It is suggested that the concentration of copper in the ovum at maturity may have something to do with limiting the growth of the egg; that the elimination of copper-bearing pigment during sats ict at eta tema eames acl eal COPPER, ENZYMES, AND FERTILIZATION. 103 fertilization may indirectly restore or produce conditions essential for further growth and development. XIV. LITERATURE. Aron, H. 709 ©=Die Anorganischen Bestandteile der Tierischen Substanz. Oppenheim- ers Handb. d. Biochemie, Vol. I., pp. 62-90. Bayliss, W. M. 7152 Principles of Gen. Physiology. Longmans Green & Co., London, 1915. Behrens, H. 794 A Manual of Microchemical Analysis. MacMillan & Co., London, 1894. von Euler, H., and Svanberg, 0. "202 Ueber Giftwirkungen bei Enzymreaktionen. I. Inaktivierung der Sac- charase durch Schwermetalle-Fermentforschung, Vol. 3, pp. 330-393. Fleurent, E., and Levi, L. ’tI01 Chemical Abstracts, Vol. IV., p. 3466. Glaser, 0. "131 On Inducing Development in the Sea Utchin (Arbacia punctulata) together with Considerations on the Initiatory Effect of Fertilization. Science, Vol. XXXVIII., pp. 446-450. "14 A Qualitative Analysis of the Egg-Secretions and Extracts of Arbacia and Asterias. Biot. Butu., Vol. XXVI., pp. 367-386. 15 Can a Single Spermatozoon Initiate Development in Arbacia? Brot. Butt., Vol. XXVIII., pp. 149-153. 213 Note on the Pigment in Arbacia Egg-Secretion. Brot. Butt., Vol. XLI., pp. 256-258. 214 Fertilization and Egg-Secretions. Biot. Butt., Vol. XLI., pp. 63-72 215 The Duality of Egg-Secretion. Am. Naturalist, Vol. LV., pp. 368-373. 221 The Temporary Concentration of Sea-Salts about Arbacia Eggs Brot. Burz., Vol. XLIII., pp. 175-183. 222 The Hydrolysis of Higher Fats in Egg-Secretion. Brot. Butt., Vol. XLIII., pp. 68-74. 223 Note on the Synthesis of Ethyl Butyrate in Egg-Secretion. Science, Vol. LV., p. 486. Lillie, F. R. - 7132 The Mechanism of Fertilization. Science, Vol. XXXVIIL., pp. 524-528. 1g Problems of Fertilization. Univ. Chicago Press. 1919. ’o11 The Initial Event in Fertilization. Anat. Record, Vol. 20, p. 225. 212 Studies of Fertilization, X. The Effects of Copper Salts on the Fertilization Reaction in Arbacia and a Comparison of Mercury Effects. Brot. Butt., Vol. XLI., pp. 125-143. Macallum, A. B. "12 Die Methoden der Biologischen Mikrochemie. Abderhalden’s Handb. d. Biochemischen Arbeitsmethoden, pp. 1099-1147. Maquenne, L., and Demoussy, E. %208 Sur la distribution et la migration du cuivre dans les tissues des plantes vertes. Compt. Rend., Vol. 170, pp. 87-93. 104 OTTO GLASER. Moore, B. ’207 Biochemistry. Longmans Green & Co., N. Y. 1921. Muttkowski, R. H. ’216 Copper: Its Occurrence and Ré6le in Insects and other Animals. Anat. Record, Vol. 20, p. 230. Nageli, C. von. 793. «Ueber Oligodynamische Erscheinungen in Lebenden Zellen. Deutsch. Schweiz. Naturforsch. Gesell., Vol. 33, 1. Rose, W. C., and Bodansky, M. ’201 Chemical Abstracts, Vol. 14, p. 3436. Titze, C., and Wedemann, W. ’r11 Chemical Abstracts, Vol. V., p. 2504. Woodward, A. E. 718 Studies on the Physiological Significance of Certain Precipitates from the Egg-Secretions of Arbacia and Asterias. Journ. Exp. Zodl., Vol. 26, PP. 459-501. Yagi, S. ’t02 Chemical Abstracts, Vol. IV., p. 1761. (OAR Bind Bled ~ amarine Btological Laboratory ~ WOODS HOLE, MASS. eR Marci, 1933 8 fe Nb. 3 ee . o : a CONTENTS AR 141923. ete; i Not dat an ‘ Witte Differ ence 2 m Terms of an Environmental One ae tos, a ZELENY, Cnantes, ve he Temperature Coefficient of aye Expression for ihe Value of a EBs : Stidies on ihe Physiology of Recomsinnuleor im Plan- > aria lata, with a Description of the Species Sas PUBLISHED MonTBLy BY THE MARINE BIOLOGICAL LABORATORY - PRINTED “AND” ISSUED BY ; _ THE NEW ERA PRINTING COMPANY, ‘Tye. FES RE SOE ANCASTER BA. AGENT FOR GREAT BRITAIN - WHELDON & WESLEY, Limitep 2,3 and 4g Arthur Street, New Oxford Street, London, W. P.2_ “Single Numbers, 75 Cents. Per Volume (6 numbers), $3.00. f : Ge i ‘ mi ) { {; 1 "i ns s \ } AN si if : iy . rs | AAA Editorial Stat Ceo i yt a SY = E. e ConKiin— Princeton University. _Grorce T. Moore— Zhe Missouri Botanic Garden. Mt T; Hi ‘Morcan— Columbia University. Ww. M. Wuerter— Harvard University. y ue ; a ele eam NN x : ek | op B. Witson— Columbia University. | (rn “@panadinto Editor : # RANK R, Livrize— Zhe University of Cheng. All communications and manuscripts should be sent to the Man- aging Editor, the University of Chicago, Sept. 15th to June 15th, or. Woods Hole, Mass., June 15th to Sept. 15th. Subscriptions and other matter should be addressed to the” Biological Hee Prince and yy Lemon Streets, Lancaster, Pare reas Peed ates ata Vol. XLIV. March, 1923. No. 3. mOLOGICAL BULLETIN PEt hE NPE ATURE \COLPEICIEND 7On A HETERO— ZVECOm ER Wh AN EX PRESSFON ROR: DH, VALUE OF A GERMINAL DIFFERENCE IN TERMS OF AN ENVIRONMENTAL ONE. CHARLES ZELENY. It has been shown by Seyster and Krafka that the size of the eye and the number of its ommatidia in “bar-eye” Drosophila varies with the temperature at which the larve are developed. An increase of one degree Centigrade produces on the average a de- . crease of about 10 per cent. in ommatidial number. In ultra-bar, an allelomorph of bar and full, the change is about 8 per cent. per degree. On the other hand, full eye has a much lower temperature coefficient. Counts being made at present by Miss Karrer show a change of only 2% per cent. per degree. Since the effect upon bar and ultra-bar is so much different from that upon full, it becomes a matter of interest to determine the reaction of the heterozygotes. Are they intermediate in this re- spect as well as in ommatidial number? The results may be ex- pected to throw some light upon the manner of reaction of the genes and on the nature of dominance. The present report deals with the ultra-bar heterozygotes. The effect of temperature upon ultra-bar homozygotes has been determined by Krafka (1920, p. 416). His values are copied in the next to the lowest line of Table I. From these values the average effect of one degree of change in temperature may be determined as follows: The average ommatidial number is 51.5 at 15° and 15.8 at 30°. The difference between the logarithms of 51.5 and 15.8 divided by fifteen and reduced to its arithmetical 1 Contribution from the Zodlogical Laboratory of the University of Illinois. No. 2to. 105 106 CHARLES ZELENY. value is 1.080. In other words, one degree decrease in tempera- ture between 30° and 15° causes an average increase of 8 per cent. in ommatidial number. Between 27° and 15° the average increase is 7.6 per cent. F B ray U § Fic. 1. F, full eye. B, bar. U, ultra-bar.. Hyp, heterozygote of full and bar. Hpyy, heterozygote of full and ultra-bar. In some work not yet published, but which she has allowed me ‘to use for the present purpose, Miss Karrer finds that full eye has an average increase of only about 24 per cent. per degree between 29° and 15>. Taste I. Number of Ommatidia. 15; 20. 27. 30. Two full eye factors. Homozygous PUTO Ves scien ayes vegent aoeenes enteral eo I,077.5 947.0 810.6 754.9 One full eye and one ultra-bar factor. Weterozygote sai: sik ds cten oe eo etee II2.1 72.1 40.5 37.1 Two ultra-bar factors. Homozygous WEG ASD AD as scans seesstesee ual ate eee 51.5 32.6 Qing 15.8 Difference between heterozygote and : homozygous ultra-bar............ 60.6 39.5 I9.2 21.3 TEMPERATURE COEFFICIENT OF A HETEROZYGOTE. 107 Heterozygotes Exhibst a Complete Change in Reaction from the Full to the Ultra-bar Type.—In the case of heterozygous females, containing one full and one ultra-bar factor, the mean ommatidial values as determined by the present experiments are given in the fsccondline vot Rable 1. There are) 37.1 omimatidia at 30°; Aor at 27-, and 112.1 at 15°. ‘The average inerease per degree determined by the method previously described is 7.7 per cent. between 30° and 15° and 89 per cent. between 27° and 15°. These values are strikingly different from 2.5 per cent., the value for full eye, and strikingly similar to 8.0 and 7.6, the values for homozygous ultra-bar. The temperature coefficient of the heterozygotes must, therefore, be considered as essentially like that of the ultra-bar parent and ‘wholly different from that of the full parent. While a single ultra-bar factor is not sufficient to bring about the complete effect in reducing ommatidial number as produced by two ultra-bar fac- tors, it is sufficient to produce the complete change to a physiologi- cal system of the ultra-bar type. The change in ommatidial number with change in temperature can be explained most satisfactorily by assuming a differential effect of temperature upon the physiological processes involved in ommatidial production as opposed to other physiological processes. In view of the fact that temperature is effective only during a few hours of larval life, it may be considered that the initial steps in the formation of ommatidia are confined to a definite embryologi- cal period. The length of this period is determined by the general physiological processes of the larva, while the rate of formation of ommatidia during the period is a function of special processes which have a different coefficient. It is evident that under these circumstances two different temperatures must give two different ommatidial numbers. The difference in the temperature coeffi- cients is slight in full eye and the mutation to bar or ultra-bar involves a marked increase of this difference. Further analysis awaits a more accurate knowledge of the nature of the embryo- logical processes involved. Whatever the character of these proc- esses, however, it is clear that the reaction system produced by the introduction of a single ultra-bar factor is of the ultra-bar type, even though the reduction in ommatidial count at any specific tem- 108 CHARLES ZELENY. perature is not as great as that produced by two ultra-bar factors. © The analysis makes it evident that there are two distinct proc- esses involved in the mutation from full to ultra-bar. One of these consists in an essential change in the type of reaction as shown by the change in the temperature coefficient. The other process involves a change in the general level of the rate of the reaction without affecting its specific character. ‘The first is fully accomplished by a single ultra-bar factor. The second is influ- enced quantitatively by the number of ultra-bar factors. The Effect of Temperature upon Dominance.—Since the hetero- zygote has the same temperature coefficient as homozygous ultra- bar, and one that is much greater than that of homozygous full eye, it becomes a matter of interest to consider the effect of tem- perature upon dominance. Elsewhere (1920, p. 308) I have dis- cussed a method of determination of the coefficient of dominance by the use of a factorial scale in which the effect of a degree of temperature is taken as the measure of an unit factor. It is obvious that on this basis there can be no change in dominance with temperature, because the ommatidial value of the unit varies with change in effect of temperature. If, however, it is desired to get an expression for dominance which is based directly upon the somatic expression—1.e., upon the ommatidial number—such a value changes with the temperature. Suppose that complete or 100 per cent. dominance of full is a con- dition in which the heterozygote has the same ommatidial number _as full eye and complete recessiveness or zero per cent. dominance of full a condition in which the heterozygote has the same omma- tidial number as ultra-bar. Likewise suppose that the ommatidial count is the scale of values. Then the coefficient of dominance of full as expressed on a percentage basis is H—B'u Pla, X 100, ° in which H is the ommatidial count of the heterozygote, B’u that of ultra-bar, and-F that of full. Correspondingly F—H Dib) ee CD BY a meg CDi TEMPERATURE COEFFICIENT OF A HETEROZYGOTE. 109 In this way the coefficient of dominance of full eye is readily deter- mined as 5.9 per cent. at 15°, 4.3 per cent. at 20°, 2.4 per cent. at 27°, and 2.9 per cent. at 30°. The general decrease with increase in temperature is evident. The reversal between 27 and 30 may probably be explained by the disturbance resulting from an ap- proach to the maximum temperature. The Value of a Germinal Factor in Terms of an Environmental One.—Perhaps the most interesting point in connection with the present data is the demonstration that they furnish of the fact that the gene, ultra-bar, has the same type of reaction as a temperature difference. It is possible to state the effectiveness of particular germinal factors in terms of the corresponding effects of tempera- ture. Such an attempt has been made in previous studies of the bar races and the temperature coefficient has given the basis for the evaluation. Since these previous studies had shown that a change of one degree in temperature produces a change of approxi- mately 10 per cent. in ommatidial number, a “ unit” factor was, for convenience, taken as one that produces the same change. The units in the factorial scale, whether environmental or germinal, are thus expressed on the same basis. First of all, it will be well to take up the demonstration of the fact that the particular germinal difference represented by the addition of a second ultra-bar factor in place of the full factor of the heterozygote does not correspond to a constant somatic expres- sion. The difference between heterozygous and homozygous ultra- bar is not represented by a constant difference in ommatidial num- ber. Thus at 15 degrees the difference is 60.6 ommatidia, at 20 degrees 39.5, at 27 degrees 19.2, and at 30 degrees 21.3. These data are given in Table I. and in graphic form in Fig. 2, where the lengths of the heavy vertical lines are proportional to the omma- tidial differences at the various temperatures. The marked change with temperature is obvious, though the germinal difference re- mains constant. The ommatidial difference can not, therefore, serve directly as a measure of germinal difference. If, however, the unit of measurement is the effect produced by a degree of change in temperature, 8 per cent. in this case, and the effect produced by the substitution of the second ultra-bar factor for the full factor of the heterozygote is measured in terms of this IIo CHARLES ZELENY. unit, the result obtained is shown in Table II. and graphically in Fig. 3. In the latter the scale at the left represents the logarithms ommatidial means 120 110 100 90 80 70 60 50 40 30 20 10 0 temperatures 15° 20° A fs 30° ommatidial 60.6 Eau differences We) 52 21.3 Fic. 2. The length of each heavy vertical line represents the difference in ommatidial number between homozygous and heterozygous ultra-bar at the respective temperature. The numerical values are given at the bottom of the figure. The heterozygotes have one full and one ultra-bar factor. The scale at the left represents ommatidial numbers. The marked change ‘in the ommatidial difference with change in temperature is to be noted. Taste II. FACTORIAL VALUES. Eight Per Cent Units. 15°. 20° 270 30° One full eye and one ultra-bar factor. | _ eterozygote. s.. i's eee sce decree 25.3 +19.6 +12.3 -+II.0 Two -ultra-bar factors. Homozygous 5 ce UH Tal= DENIGP BE ie, cack ac ees oe Arms ore ae Oe +15.3 + 9.4 + 3.9 0.0 Difference in eight per cent. units or ee: temperature units. .os....5..2..- = 10.3 10.2 8.4. II.0 TEMPERATURE COEFFICIENT OF A HETEROZYGOTE. III of the ommatidial counts, arranged so that each unit has the same logarithmic value, corresponding to a change of 8 per cent. in ommatidial number. ‘The location of the zero point is of course purely arbitrary. logs of scale of ommatidial degrees means 2.034 +25 1.868 +20 1.700 +15. 1.533 +10 1.366 +5 - 1.199 ka temperatures 15° 20° 27° 30° log. differences 0.338 0.345 0.279 0.971 Fic. 3. The length of each heavy vertical line represents the difference in terms of logarithms of ommatidial number between homozygous and hetero- zygous ultra-bar at the respective temperatures. The numerical values are given at the bottom of the figure. The heterozygotes have one full and one ultra-bar factor. The logarithmic scale is represented at the left. In the next column is the scale of corresponding degrees Centigrade which would produce the same effect, starting with an arbitrary zero at the lowest value of the present observations. It is to be noted that the difference between homozygote and heterozygote on this scale is fairly constant. The heavy vertical lines again give the effect produced by the substitution of a second ultra-bar factor for the full factor of the heterozygote at the different temperatures. The second ultra-bar factor at 15° C. depresses the value by an amount equal to that produced by 10.3 degrees of temperature, at 20° C. by 10.2 de- grees, at 27° C. by 8.4 degrees, and at 30° C. by 11.0 degrees. The average depression is the same as that produced by ten de- grees of increase in temperature, and considering the character of I1I2 CHARLES ZELENY. the determinations the values are remarkably constant. The con- clusion may be safely made, therefore, that a proper measure has been found for the expression of the germinal value in question. A homozygous ultra-bar at 15° has the same ommatidial number as a heterozygote at 25° and a homozygote at 20° the same number as a heterozygote at 30°. On the logarithmic scale of ommatidial number it is, therefore, possible to express the relation between the germinal factor and the environmental factor as a constant. The present data, there- fore, strengthen the validity of the use of this scale as a measure of the germinal factors as well as the environmental ones. LITERATURE CITED. Krafka, Joseph. ’20 The Effect of Temperature upon Facet Number in the Bar-eyed Mu- tant of Drosophila. J, Gen. Physiol., 2: 409-464. Seyster, E. W. ; "19 Eye-Facet Number as Influenced by Temperature in the Bar-eyed Mu- - tant of Drosophila melanogaster (ampelophila). Biot, BuLL., 37: ; 168-182. Zeleny, Charles. 720 A Change in the Bar Gene of Drosophila melanogaster involving Fur- ther Decrease in Facet Number and Increase in Dominance. J. Exp. Zool., 30: 203—324. "22 The Effect of Selection for Eye-Facet Number in the White Bar Eye Race of Drosophila melanogaster. Genetics, 7: I—115. ee ae Sew ON) Tiaik PHYSIOLOGY On RECONSTITU— MON GUINEAN A LATA Ri A DESCRIP — MON OF THE Siri Gils: P, B. SIVICKIS, Hutt ZoodtocicaL Laporatory, UNIVERSITY OF CHICAGO. The earlier work on the reconstitution of isolated pieces of planarians into new individuals was very largely concerned with the description of the visible changes, such as the outgrowth of new tissue, its differentiation, the reorganization and redifferenti- ation of the old parts, the changes in shape, the minimum size of pieces capable of reconstitution, and the relations between the polarity of the new individual and that of the animal from which the piece was taken. Most of the experiments were performed with small numbers of pieces, often from different regions of the ' body and without any attempt at a physiological standardization of the experimental material. During the past twenty years Planaria dorotocephala has been extensively used in this laboratory as mate- rial for experimental investigation, and in the course of this work various methods have been developed which have made possible some degree of physiological analysis of the process of reconstitu- tion in this species. In this work physiological standardization of material, mass experiments, and control of environmental condi- tions have played an important part. The desirability of including other species within the program of investigation has become increasingly evident with the progress of the work on P. dorotocephala, and at the suggestion of Dr. C. M. Child the analysis of reconstitution in a closely related species was begun. The present paper comprises a part of the results of this investigation. In this connection I take the opportunity to acknowledge my deep indebtedness to Professor Child for his advice, friendly criti- cism, and revision of the manuscript. I also wish to express my thanks to Dr. L. H. Hyman for the data on oxygen consumption presented in this paper and for many helpful suggestions. 113 114 P. B. SIVICKIS. MATERIAL: DESCRIPTION OF SPECIES. In many rivers and lakes about Chicago a planarian, commonly identified in the past as P. maculata, occurs in large numbers. Even a cursory examination makes it evident that this form differs in various respects from P. maculata of the Atlantic slope. Atten- tion has already been called to these differences by Hyman (20). A comparison of living individuals of this form and P. maculata from the region of Woods Hole, Mass., shows the following differ- ‘ences: The pigment pattern of the mid-western form (Fig. 1) is distinctly coarser and more irregular ; the individual pigment spots and the unpigmented areas are more clearly visible to the naked eye than in P. maculata (Fig. 2). In a mixed stock the two ae PEPE teat Went Fic. 1. Planaria lata n. sp. Fic. 2. Planaria maculata. PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. II5 forms are at once distinguishable by these differences in pigmenta- tion. Moreover, the general color effect to the naked eye in the mid-western form is a light grayish brown, mottled or dappled, while in P. maculata the brown tint is much deeper and more uni- form. In P. maculata a light median longitudinal stripe is almost invariably present (Fig. 2), while the mid-western form usually ’ shows an obscure dark median stripe (Fig. 1). As regards shape of body, the mid-western form is distinctly broader in proportion to length than P. maculata (cf. Figs. 1 and 2). This difference is evident during locomotion as well as at rest. The cephalic lobes are apparently slightly less developed and the digestive tract appears more highly branched in the mid- western form than in P. maculata. As regards motor behavior, also marked differences appear. The mid-western form is distinctly more sluggish, reacts more slowly, and progresses less rapidly than P. maculata. Similar differences in reaction to food exist. P. maculata can be collected by placing pieces of meat in the water in localities where the species occurs. As the extractives diffuse, the animals will come to the meat from a distance of several inches in standing water and ~ from greater distances in flowing water. The mid-western form can not be collected in any considerable numbers in this way be- cause only those animals immediately about the meat react. In feeding stocks in the laboratory it has been found necessary to grind the meat and spread it over the bottom of the container, instead of placing pieces at intervals of two or three inches, as has been the practice with P. dorotocephala and P. maculata. All these differences persist unchanged in stocks of the two forms kept in the same water and under the same conditions in Chicago. This is true, not only as regards the original individuals, but also as regards young animals resulting from fission or recon- stitution of pieces and animals hatched from eggs. Reconstitu- tion experiments with the two forms also show certain character- istic differences in head frequencies. These differences are unquestionably sufficient to make it evi- dent that the mid-western form is not P: maculata. However, since the morphological characteristics of the organ complex of the genital atrium, and particularly the copulatory organ, are com- 116 P. B. SIVICKIS. monly regarded as the most trustworthy criterion of specific differ- ences, sections of this region of sexually mature animals have been made. 4 Fics. 3, 4. Atrial genital complex of the two species of Planaria: Fig. 3, P. lata; Fig. 4, P. maculata combined from two adjoining sections; c, copula- tory organ; f, female region of atrium; g, circular groove or furrow on copu- latory organ; m, male region of atrium; wv, valve or fold between the two parts of atrium. Fig. 3 is a median sagittal section through this region of the mid-western form and Fig. 4 a similar section from P. maculata, both from sexually mature animals. The circular groove (g) about the apical region of the copulatory organ (c) is deeper in P. lata and the outline of the fold or valve (vw) between the com- mon portion of the atrium and the female duct (f) is very differ- ent in the two species. Other differences in shape of the parts and cavities are evident, but may be due in part to differences in muscular contraction. The differences in the atrial complex con- firm the conclusion that the two forms are different species, and since the mid-western form does not agree with descriptions of other species already given, it is evidently an unnamed species and is named and described as follows: Planaria lata n. sp—Length of full-grown, sexually mature in- PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 117 dividuals averaging 12-14 mm., occasionally 16 mm. Body rela- tively broad, in full-grown animals width from tip to tip of ce- phalic lobes and from margin to margin at mouth about one sixth of length. Pigment pattern coarse, irregular, often with obscure dark median stripe, the general effect being mottled or dappled light grayish brown. Cephalic lobes short. Animal sluggish and markedly less sensitive to external factors than either P. doroto- cephala or P. maculata. Structure of organ complex of genital atrium as in Fig. 3. The present paper is primarily concerned with this species, but attention is called to the physiological differences between this species and P. dorotocepha which have been brought to light by the experiments. THE OCCURRENCE OF FISSION AND SEXUAL MATURITY. Like P. dorotocephala, P. lata shows no visible morphological - indication of the presence of a posterior zodid, but, as will appear below, the presence of such a zooid can be demonstrated by physi- ological methods. Fission in P. maculata has been described by Curtis (02) and the act of fission has been observed in P. doroto- cephala by Child (10, ’11c). In the latter species it consists in an independent motor reaction of the posterior zodid while the animal is moving forward. ‘The posterior zooid attaches itself, the ante- rior zooid continues to advance, and the body in front of the attached portion is finally ruptured. Fission is much more likely to occur after slight stimulation than after violent disturbance, for in the latter case the posterior zooid is controlled by the anterior and does not react independently. The act of fission has not been observed in P. lata, but undoubtedly it occurs in the same manner. In the laboratory fission is often induced by changing the water, but does: not occur at once, while the animals are very active, but only later as their activity decreases. As regards the level of the body at which fission occurs, P. lata differs markedly from P. dorotocephala. In the latter species fis- sion normally occurs at a level 1-3 mm. posterior to the mouth (Fig. 5), and in cases of delayed fission in the laboratory the posterior fission piece may be longer than the anterior. In P. Jaa, however, fission takes place much nearer the posterior end (Fig. 6), 118 P. B. SIVICKIS. and in animals 12-13 mm. in length the posterior fission piece is only 2-3 mm. in length, and in shorter animals it is not only abso- 5 Fic. 5. P. dorotocephala, showing level at which fission usually occurs. Fic. 6. P. lata, showing level at which fission usually occurs. lutely, but relatively, shorter. In P. dorotocephala the posterior fission piece often divides again after four or five days, when it begins to move about more or less normally. Apparently at this stage the developing head is unable to control the whole length of the posterior fission piece and fission takes place at one of the more posterior zodid boundaries. Such second fission has never been PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 119 observed in P. lata, but the susceptibility (p. 32) and the head frequency (p. 37) of the region just anterior to the level of fission suggest that some slight degree of physiological isolation and the earliest stages of another zodid exist there in large animals. If such a zooid is present, it develops to a stage at which fission is possible only after fission has occurred posterior to it and it has become the posterior end of the body. The maximum length attained by P. Jata under ordinary condi- tions in nature and in the laboratory is twelve to fourteen milli- meters, and at this size thé animals become sexually mature, even though fission continues to occur. The level of fission is so far posterior to the genital pore that the development of the ducts and pore is usually not affected to any appreciable degree by fission. In this respect also this form differs from P. dorotocephala, in which the level of fission is so near the level of the genital pore that the occurrence of fission in an animal approaching sexual maturity usually brings about disappearance of the pore and at least the posterior portions of the ducts. In the localities about Chicago P. lata becomes sexually mature and deposits eggs from June to September. In the laboratory maturity and egg laying may occur at any time of year. In P. dorotocephala sexual ma- turity may occur in the laboratory when the animals are well fed and fission is prevented by keeping them on slimy or vaselined surfaces, but it has not been observed under natural conditions in this region, being apparently prevented by fission and perhaps also by periodic starvation (Child, ’11c). Curtis also found that in some localities P. maculata does not become sexually mature, but did not discover the determining conditions. It may be suggested that an environment which inhibits slightly the physiological ac- tivity of the animals and so decreases the range of dominance may result in the occurrence of fission at a more anterior level, and this may interfere with the development of the genital ducts and pore. Experiments-to test this suggestion have not yet been performed. MEeETHODS. The animals are found both on stones at the bottom and on Elodea and other water plants at various levels. On the stones all sizes from very young to sexually mature animals and numerous I20 P. B. SIVICKIS. egg capsules may be found in summer, but among the plants only the smaller, younger animals have been collected. Stocks are kept in the laboratory for three or four weeks before using experi- - mentally so that uniformity in nutritive conditions may be approxi- mated. They are fed three times a week with ground and washed beef liver, as described by Hyman (’20). The work in this laboratory with planarians has demonstrated that in a stock of animals collected at one time, kept under as nearly as possible identical conditions of temperature, nutrition, © water supply, etc., size is the best criterion of physiological condi- tion, and particularly of physiological age, as indicated by suscepti- bility and respiratory rate, which can readily be applied to the liv- ing animals in the selection of material for experiment. Unin- jured animals of the same size from such a stock show a high degree of uniformity in susceptibility to chemical and physical agents and in rate of respiration, as shown by the work of Child, Hyman, Behre, and Buchanan, and are more alike physiologically than material selected on any other basis thus far discovered. In these animals the amount of growth, whether rapid or slow, and consequently the size of the individual, is a far more exact measure of their physiological age, and so of their susceptibility and rate of respiration, than the length of time they have lived as indi- viduals (Child, ’15a, Chap. [V.; Hyman, ’19 C). Since the experiments recorded in this paper are all mass ex- periments—i.e., with numbers of individuals—the material for each experiment is selected on the basis of size. Such standardization of material is necessary for the attainment of definite results which can be predicted and controlled and it makes possible pre- diction and control to a high degree. In the course of the work experiments were performed with standardized material from general stocks collected and kept as described above, from stocks composed of animals hatched from eggs in the laboratory, and from stocks grown from cut pieces. The experimental data presented below concern chiefly respira- tion, susceptibility, and head frequencies—+z.e., the frequencies of occurrence of the various forms of head in the reconstitution of pieces in relation to level of body, length of piece, and physiological age of animals. Some experiments on modification and control of ES PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. ILA head frequency by means of chemical agents have been performed, but are only briefly mentioned. The data on respiration include comparative estimations of CO, production by colorimetric determination of pH with phenol-sul- phone-phthalein as indicator and the indicator-buffer solutions made up by Hynson Westcott and Dunning as standards. In these experiments lots of as nearly as possible equal weights of animals or pieces to be compared are sealed in pyrex tubes of the same diameter as the standard tubes in equal volumes of indicator solution of the same concentration as the standard tubes and the change in color recorded at regular intervals and usually also to a definite pH. Some data on oxygen consumption determined by the Winkler method are also given. For these I am indebted to the kindness of Dr. Hyman. In the experiments on susceptibility KNC has been chiefly used _as agent, since it has been abundantly demonstrated that with proper precautions susceptibility to KNC can be used as a general comparative measure of physiological condition and particularly of rate of respiration in Planaria.1_ In the present study the sus- ceptibility method has been used for two purposes: first, for dem- onstration of the axial gradients and the posterior zooid by the differential susceptibility of different levels of the body; second, to demonstrate the changes in physiological condition of pieces following section. The general susceptibility gradients are shown in figures and the comparative susceptibilities of pieces by graphs. The method of graphing the data is described in connection with the data. In the experiments on reconstitution the body posterior to the head is cut into a certain number of pieces—four, six, eight, sixteen—according to the experiment. Animals of the same size are used for each experiment and the pieces from each individual are cut as nearly as possible the same length, the more extreme irregularities being discarded. Consequently the corresponding pieces from different individuals represent as nearly as possible the same region of the body. All corresponding pieces—t.e., those 1 Child, ’14a, c, ’15a, chaps. III—VII., euGe *t9a, b; Hyman, ’19a, b, c, ’20, ’22. 122 P. B. SIVICKIS. representing the same body region—are kept together in one con- tainer. The usual number of pieces in each lot is fifty, each taken, of course, from a different individual of the size used in that ex- periment. The pieces are allowed to undergo reconstitution and examination is made and results recorded after twelve to fourteen days, at which time reconstitution is so far advanced that no fur- ther change in the character of the form produced will occur. Since the forms produced fall into certain groups or types, as described in a following section, the results can be tabulated to show the frequency of each type in each lot. In this species, as in P. dorotocephala, the head is the most characteristic distinguish- ing feature of the different forms produced, except in very short pieces. The basis of tabulation is, therefore, form and structure of the anterior end, and the tables show the frequencies of the different forms and are called, for convenience, head frequency tables, this term having been used for similar data on P. doroto- cephala. From the tabulated head frequencies graphs are con- structed by a method described below, and in this way the head frequencies in different regions, in pieces of different lengths, and in animals of different size or physiological condition may be directly compared. The relation of head frequency to region of body and length of piece is shown by comparing pieces of different lengths and from different body levels of animals of the same size, and therefore approximately of the same physiological age. The relation of head frequency to physiological age is determined by comparing the results obtained with pieces from animals of different lengths. In my experiments animals of two standard lengths have been chiefly used: full-grown animals, 11-13 mm. long, physiologically old, and with low rate of respiration; and young growing animals, 4—6 mm. long, with much higher rate of respiration. The smaller size is the smallest which can be conveniently used for such experi- ments because of the difficulty of cutting pieces of equal size in the smaller individuals. Some experiments have been performed with sizes intermediate between these two extremes, some with animals raised from eggs and some with animals raised from cut pieces. PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 123 THe AXIAL SUSCEPTIBILITY GRADIENTS. The occurrence of definite axial susceptibility gradients in both animals and plants and their relation to metabolic rate has been discussed in various publications (e.g., Child, ’20b). Various lines of evidence show that susceptibility to a wide range of chemi- cal agents in concentrations or intensities above the limit of toler- ance or acclimation and below the limit at which death occurs immediately varies in general directly with metabolic rate, or more specifically with rate of respiration. Susceptibility gradients are characteristic features of physiological axes in both plants and animals and many facts indicate that such axes are primarily quantitative physiological gradients. It has been found that in P. dorotocephala the head frequencies at different levels of the body show a definite relation to the polar susceptibility gradient, regions of high susceptibility being regions of high head frequency and vice versa. The observations on sus- ceptibility in P. lata show that a similar relation exists in this species. In these observations KNC was used in most cases as agent, because it is known to be a powerful inhibitor of oxidations and has been extensively employed in the study of susceptibility, but many other agents of very different constitution—e.g., HgCl., CuSo,, acetic acid, various anesthetics, etec—in proper concentra- tion give essentially the same results. The procedure consists, first, in determining a concentration which kills slowly enough to permit the differences in susceptibility to appear clearly and which is not low enough to permit acclimation, and, second, in observing the progress of disintegration of the tissues of animals placed in the solution, in closed containers if necessary to avoid loss from volatilization. As death of any part occurs or approaches struc- tural disintegration of the tissues takes place, and such disintegra- tion appears first in certain regions and follows a definite course, so that certain body regions are completely disintegrated while others are still intact and moving. The general course of disintegration in P. lata is shown in Figs. 7-10. The head and the posterior zooid are most susceptible and disintegration progresses posteriorly from the head and at the same time involves a region anterior to the posterior zodid and 124 P. B. SIVICKIS. posterior to the genital pore, the last region to disintegrate being that between the mouth and the genital pore. Commonly the lat- eral margins of the body disintegrate somewhat earlier than median regions at the same level, but in some individuals these transverse differences do not appear, and in some the median region appar- ently disintegrates earlier than the margins. There is not much difference between dorsal and ventral, but ventral regions usually disintegrate slightly in advance of dorsal (Fig. 11). Fics. 7-11. Susceptibility gradients in P. Jata, as shown by KNC m/1000: Figs. 7-10, stages in progress of disintegration in dorsal view; Fig. 11, a stage of disintegration in lateral view to show difference between dorsal and ventral, It should be noted that these statements concern primarily the body wall, but they appear to hold good for the parenchyma also. In well-fed animals the digestive tract is highly susceptible and, even though the cyanide must pass through the body wall to reach it, the digestive tract usually swells and disintegrates earlier than the body wall and often bursts through the latter at various points. In starved animals the digestive tract is much less susceptible and PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 125 in advanced starvation may remain intact longer than the body wall. In general, the course of disintegration in this species is very similar to that in P. dorotocephala, but some minor differences appear. In long individuals of the latter, possessing more than one zooid in the posterior region, the different zodids usually show different susceptibilities (Child, ’t1c,.’°13b). The early disinte- gration of the region anterior to the posterior zodid in P. lata may indicate the presence here of another zooid at a very early stage; in other words, some degree of physiological isolation with increase of metabolic activity may exist anterior to the level of fission and yet be insufficient to permit fission at this level. In P. dorotocephala the margins of the body always disintegrate in KNC before median regions, while in P. lata this may or may not be the case. These differences are apparently associated with the specialization of the margins as motor organs, and particularly as organs of secretion of slime. There are many glands in this region and the alkaline cyanide stimulates these to increased ac- tivity. Often the separate glands disintegrate before other parts of the margins, particularly in P. dorotocephala. Apparently the margins of P. lata are less highly specialized in this way, for the glands appear less distinctly as regions of disintegration and there is less motor activity of the margins. In neutral or acid cyanide and in other acid agents the glandular activity is apparently not increased and motor activity is decreased, and in such solutions the susceptibility of the margins, even in P. dorotocephala, is usu- ally less than that of median regions. In P. dorotocephala the dorsal body wall disintegrates earlier than the ventral in alkaline cyanide. The dorsal wall is thinner than the ventral and shows many localized regions of disintegra- tion, apparently glandular. In acid agents ventral regions are usually more susceptible. In P. lata the specialization of the dorsal surface is apparently less, so that even in alkaline agents the differ- ence between dorsal and ventral is slight and the ventral surface is usually the more susceptible. In the early developmental stages of turbellaria the median ven- tral region is more susceptible than lateral and dorsal regions (Child, unpublished), and in full-grown planarians the new tissue 126 P. B. SIVICKIS. develops more rapidly from the median ventral region than from other parts of a cut surface. This fact suggests that internally the median ventral region still possesses the highest metabolic rate. Apparently the primary symmetry gradients undergo more or less alteration in the body wall of some species, and the susceptibility data indicate that such alteration is greater in P. dorotocephala than in P. lata. The latter species seems to retain more nearly the characteristics of earlier stages. Young animals are always more susceptible than old and the differences in susceptibility at different levels of the body are less in the young. In fed animals susceptibility decreases from the time of hatching to maturity and in this respect parallels rate of respiration (Child, ’19a; Hyman, ’19c). In full-grown animals the time from the beginning to the end of disintegration in m/1,000 KNC at 20° C. is eight to ten hours; in young animals it is much less, increasing with advancing age. High temperature increases, low temperature decreases susceptibility. Starvation increases susceptibility and also decreases the differences at different levels. In all these respects the two species are alike. Tue ForMs RESULTING FROM RECONSTITUTION. As in P. dorotocephala, the results of reconstitution differ in definite and orderly ways according to length of piece and region of body from which it is taken. Some of these differences, such as the level at which the new pharynx appears, length of pre- pharyngeal and postpharyngeal regions, are merely temporary fea- tures of the earlier stages of reconstitution, and are later largely or completely obliterated by differential growth of prepharyngeal and postpharyngeal regions, particularly if the new individuals are fed, until finally all normal or nearly normal individuals attain approximately the same proportions. The most conspicuous differences in the results of reconstitution concern the head and these differences are, with certain rare excep- tions, permanent. Isolated pieces do not always develop anterior ends like that of the normal animal in nature, but abnormal forms occur which constitute a continuous series with some secondary modifications from the normal head to a completely headless con- ee ee ee PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. L277, dition. The members of this series fall readily into the same types or groups as in the case of P. dorotocephala (Child, ’11a, ’21), viz., normal, teratophthalmic, teratomorphic, anophthalmic, ace- phalic. The normal head is like the head of Fig. 1. The terat- ophthalmic head is normal in outline, but the eyes are more or less approximated to the median line and the pigment spots are often partially united or even fused (Figs. 12, 13). In the terato- morphic head there is a single, or apparently single, median eye and the anterior region between the cephalic lobes is incompletely developed, so that the lobes appear more or less anteriorly or fused at the median line (Figs. 14, 15). The anophthalmic head is merely an outgrowth of tissue without eyes (Figs. 16, 17), and in o @) ® © 19 17 18 16 A 20 Fries. 12-20. Forms resulting from reconstitution: Figs. 12, 13, teratoph- thalmic heads and eyes: Figs. 14, 15, teratomorphic heads; Figs. 16, 17, anophthalmic heads; Fig. 18, acephalic form; Fig. 19, tailless form; Fig, 20, biaxial heads. the acephalic form the anterior end simply heals over without out- growth (Fig. 18). As regards degree of development, the ce- phalic ganglia of these forms also constitute a continuous series from normal ganglia in the normal head to a rudimentary ganglion in 128 P. B. SIVICKIS. the anophthalmic, and no ganglia in the acephalic form (Child and McKie, ’11). In this species, as in P. dorotocephala, no other head forms have been observed except secondary modifications of some one of these forms in acclimation to, or recovery from, in- hibiting agents (Child, ’21) and inequalities or asymmetries in position of eyes resulting from oblique section or other incidental conditions and usually temporary. The frequency of occurrence ~ of these various forms in any lot of pieces is the head frequency of the lot, and the following experiments on reconstitution are chiefly concerned with the relations of head frequency to length of piece, level of body, and size of animal from which pieces are taken. The results of reconstitution in very short pieces require men- tion. Forms with heads, usually normal, but without posterior ends, “tailless forms” (Fig. 19) and “biaxial heads” (Fig. 20), appear rarely in sixths of large animals, more frequently in eighths, and their frequency increases with decreasing length of piece to the limit of length at which wound closure and reconstitu- tion fail to occur. Under ordinary conditions these forms are much more frequent and appear in longer fractions of the body in this species than in P. dorotocephala. In the latter species they have never been seen in sixths or eighths, except rarely in eighths from small young animals, and they are rare even in sixteenths and twentieths of large animals. HEAD FREQUENCIES IN RELATION To LENGTH oF PiEcE, LEVEL oF Bopy, AND PHYSIOLOGICAL AGE OF ANIMAL. The data presented in this section include head frequencies of fourths, sixths, eighths, and sixteenths of full-grown animals II-13 mm. in length and fourths and sixths of young animals 4-6 mm. in length. In the series of longer pieces the death rate is negligible, but in the sixths of young and the sixteenths of old animals it becomes high enough to lessen considerably the value of the data on head frequencies, and in pieces shorter than these. it is still higher, so that determination of head frequencies becomes impossible in such pieces. This increasing death rate with decreasing length of ieee is certainly to a considerable extent a consequence of increasing area a eo ee PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 129 of cut surface in relation to size of piece. The deaths are practi- cally limited to the first day or two following section. Pieces that survive this period live, with rare exceptions, to the end of the experiment and undergo some degree of reconstitution. In such very short pieces of P. dorotocephala the contraction of the cut surface often brings about rupture elsewhere, and contraction at this point produces further rupture in other regions and the piece gradually breaks up. In P. lata also the contraction following “PABICR le HEAD FREQUENCIES OF FULL-GRowN P. lata. (11-13 MM.) IN RELATION TO LENGTH OF PIECE AND LEVEL oF Bopy. ee Terato-| Terato-| Ano- Length of Pieces. of Body | Nor- | phthal-| mor- | phthal-| Aceph-| Dead. | Biaxial Ani- | Level. | mal. mic. phic. mic. alic. Heads. mals Fourths....... 210 A 99 — — — I — — B 47 33 Tae 6 Io 4- = Cc 17 16 = 3 3 I = D 96 I = I I I — NEXCMG aii es es 250 A 98 I — — — I — B 53 30 I 5 9 2 4 C 48 33 2 6 .8 3 — D 43 17 2 8 29 I — E 83 II I I 2 2 — F 98 I == I = — — Highths....... 850 A 95 I — — I 3 — B 413 16 5 5 6 4 6 G 46 27 I 3 16 F] 8 D Si 27 I 8 22 5 a E 39 17 2 6 32 3 I F 66 II 2 2 16 3 1.6 G 85 6 — I 4 4 9 val 96 2 a = a= 2 — Sixteenths..... 100 A 72 I — = — 27 — B 61 4 —_ ir — 34 I Cc 46 2 — 2 4 46 I D 47 8 Ta 2 8 35 3 E 39 13 = I 13 34 2 F 107] 13 — IO 22 38 2 *G 22 9 — 7 21 41 I H 29 6 ae 2 24 39 4 I 37 5 = 3 26 29 2 of 26 i = 9 28 30 3 K 34 8 = II 20 27 6 L 42 8 I 6 no) Bui 6 M 45 13 = 5 5 32 6 N 47 5 = 4 ato) 34 6 O 77 5 Sale 7 9 4 P 88 5 — = I 6 aoe 130 P. B. SIVICKIS. section is the chief factor in determining these early deaths in short pieces. In the experiments presented here the pieces dying in this manner are included in the totals in calculating percentages and in graphing, but in certain lines of future investigation it will probably be desirable to exclude these early deaths in determining total head frequencies. The head-frequency data are tabulated in percentages, the full- grown animals in Table I., the young animals in Table IJ. From ABiEle Heap FREQUENCIES OF Younc P. lata (4-6 MM.) IN ReLation To LENGTH OF PIECE AND LEVEL OF Bopy. Num- ber Terato-| Terato-| Ano- Length of Pieces. of Body | Nor- | phthal-| mor- | phthal-| Aceph-| Dead. | Biaxial Ani- | Level. | mal. mic. phic. | mic. alic. Heads. mals Hounthseeccc.) I50 A 90 5 —_ I I 3 —_— B 51 17 — I I5 16 _— Cc 72 I5 — — a 6 sal D 85 3 = I 3 8 = bide oeenony eee I50 A one 3 — — I 5 — B 61 23 I I 5 — — Cc 48 34 oa 2 9 7 = D 48 23 — I I7 It a7 E 69 6 = I 5 20 a7 1 95 4 = = I as = the tabulated data graphs are plotted by the method of assigning numerical values to the different head forms as follows: Normal heads, 5; teratophthalmic, 4; teratomorphic, 3; anophthalmic, 2; acephalic, 1; dead, o. To obtain a head-frequency value for a given lot of pieces, the number of pieces or the percentage of each head form is multiplied by the numerical value of that form and the sum of these products for the particular lot is divided by the number of pieces in the lot, or, if the data are in percentages, by one hundred. The results are the same whether actual number of pieces or percentages are used. These values plotted as ordinates against the successive levels of section, A, B, C, etc., from the heads of the original animals posteriorly as abscisse, give curves which permit direct comparison of the head frequencies of pieces of different lengths and from different levels. PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 131 Fic. 21. Head frequencies in relation to length of piece and level of body. P. lata: fourths, unbroken line, 110 animals; sixths, long dashes, 250 animals; eighths, short dashes, 850 animals; sixteenths, alternating long and short dashes, 100 animals. P. dorotocephala: fourths, dotted line, 110 animals; sixths, alternating dots and long dashes, 110 animals; ecighths, alternating dots and short dashes, 70 animals; sixteenths, two dots alternating with short dashes, 60 animals. On the outline of P. lata below the graph the levels of section of pieces of different length are indicated: A—D, large capitals, fourths; A—F, small capitals, sixths; a—h, large lower case, eighths; a-p, small lower case, sixteenths. The levels of corresponding pieces of P. doro- tocephala are in part slightly different from those indicated in the diagram since in this species the posterior zooid is mueh longer and the pharynx therefore nearer the anterior end in large animals than in P. lata. 132 P. B. SIVICKIS. = In the graph, Fig. 21, the data for full-grown animals given in Table I. are plotted, together with head-frequency curves from similar experiments with P. dorotocephala. In Fig. 22 the data for young worms from Table II. are graphed in comparison with corresponding data of P. dorotocephala.2, As regards region of body, the data show, both for full-grown and for young animals, that head frequency decreases from the most anterior level of sec- tion posteriorly and then increases again with approach to the level of the posteriod zodid, until at the most posterior level of section it is as high as at the most anterior level, except when lowered by early deaths. Second, as regards length of piece, the data show for old ani- mals that decrease in head frequency from anterior to posterior region of the first zodid and increase at levels further posterior becomes greater as length of piece becomes less. The shorter the pieces, the steeper the downward and upward slopes of the curves in Fig. 21. At the most anterior and most posterior levels of the body fourths, sixths, and eighths are practically alike in head fre- quency, sixteenths somewhat lowes anteriorly, but the differences in level of the lowest points of these curves is considerable. The irregularities in the curve of sixteenths are due to the differences in length of the pieces, the variations being, of course, relatively much greater in these very short pieces than in longer pieces. In the young worms (Fig. 22) the differences in steepness in relation to length of piece do not appear in the only data available, those for fourths and sixths, but the sixths show a somewhat lower head frequency than the fourths, except at the posterior end, where it is the same in both.? 2 The curves of head frequency of P. dorotocephala are plotted from data obtained from various sources: fourths, sixths and eighths of old animals and fourths and sixths of young animals from Child (11b, 16, ’20a), one series of eighths of old animals from Miss M. A. Hinrichs and a complete series of my own for both old and young. All of these data are in general agreement as regards relation of head frequency to length of piece, level of body and physiological age. ; 3 It may be noted that the data for young worms are made up in part from animals raised in the laboratory from eggs and in part from animals of the same size similarly raised from cut pieces. The pieces of animals from eggs showed a somewhat higher death rate and therefore a somewhat lower head frequency than those from cut pieces, but the differences were not great. t PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA, 133 | | Fic. 22. Head frequencies in relation to length of piece and level of body in young animals. P. lata: fourths, unbroken line, 150 animals; sixths, dashes, 150 animals. P. dorotocephala: fourths, dotted line, 60 animals; sixths, dots and dashes, 100 animals. The diagrammatic outline below graph indicates levels of section: A—D, large capitals, fourths ; A-F, small capitals, sixths. Since the number of animals obtained from eggs was limited, it was not possible to determine whether these differences were characteristic or merely the result of slight differences in experimental conditions. For the present, therefore, it has seemed best to combine these data in tabulation and graphing. 134 P. B. SIVICKIS. Comparison of the curves for P. Jata with those for P. doroto- cephala in Figs. 21 and 22 shows that the changes in head fre- quency with length of piece and level of body are in general of the same sort, but that their range is very much greater in P. dorotocephala than in P. lata. In full-grown animals (Fig. 21) the head frequencies of fourths of P. dorotocephala are somewhat higher than, those of sixths about equal to, and those of eighths and sixteenths far lower than the head frequencies of. correspond- ing pieces of P. lata from the posterior regions of the first zooid, while at anterior and posterior ends of the body the differences between the two species are slight. In the young animals (Fig. 22) fourths and sixths from the posterior region of the first zooid of P. dorotocephala are far below fourths and sixths of P. lata from the same region, while at the anterior end of the body the differences are much less, though greater than in old animals. The graph, Fig. 23, is a comparison of the curves of fourths and sixths of full-grown and young individuals of P. lata. This graph shows that the head frequencies in young animals are slightly lower than in corresponding pieces of old animals, but this age differ- ence is much less than in P. dorotocephala. Comparison of Figs. 21 and 22 shows that the curves of fourths and sixths of young P. dorotocephala are much steeper and fall much lower than those of fourths and sixths of old animals. Tables I. and II. show the frequencies of biaxial heads, but not of tailless forms. Unfortunately tailless forms were not recorded as such in the earlier experiments. It may be stated, however, that they do not appear in the longer pieces, that their frequency increases with decreasing length of piece, and that, so far as data are at hand, they show no definite relation to level of body. As regards the frequencies of biaxial heads, Tables I. and II. show that they do not occur in the longer pieces, that their frequency increases with decreasing length of piece, and is apparently some- what greater near the level at which fission occurs than elsewhere. EXPERIMENTAL ALTERATION OF HEAD FREQUENCY. In P. dorotocephala the head frequencies have been altered ex- perimentally by many different chemical and physical factors (Behre, 718; Buchanan, ’22; Child, ’16, ’20a).. Moreover, it has PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 135 been found that head frequencies may be altered in opposite direc- tions in pieces from different levels of the same animals by the Fic. 23. Head frequencies of fourths and sixths in old and young individu- als of P. lata: fourths, young, unbroken line; sixths, young, dotted line; fourths, old, long dashes; sixths, old, dots and dashes. The curves in this graph are taken from Figs. 21 and 22. same concentration of a single agent (Child, ’16; Buchanan, ’22). In pieces which normally show a high head frequency it may be decreased and in pieces which normally show a low head frequency 136 P, B. SIVICKIS. it may be increased by the same concentration of cyanide or anes- thetic. On the basis of these and many other facts concerning the physiology of reconstitution a theory of head frequency has been advanced (Child, ’14a, ’14d, ’16) which is confirmed by later work (Buchanan, ’22). This theory, which is considered more fully below (p. 56), maintains that head frequency in any particular ’ case is determined primarily by the relation between two opposing factors: the one the rate of metabolism in the cells at the cut sur- face which form the head, the other the rate of metabolism in other parts of the piece. The higher the first in relation to the second, the higher the head frequency; the higher the second in relation to the first, the lower the head frequency. ‘The differential susceptibility of the cells at the cut surface and the other parts of the piece makes it possible through differential inhibition, differ- ential acclimation, and recovery (Child, ’20b, ’21) to alter the relation between the two factors in both directions. The correlative factor retarding or inhibiting head formation is apparently in part or wholly a matter of nervous stimulation of the cells not directly concerned in head formation. It is most effective during the first few hours after section, when increased CO, production, oxygen consumption and susceptibility all show that the pieces are stimulated. In P. dorotocephala this stimula- tion is inhibited and head frequency increased by anesthetics such as ether and chloroform used during a few hours following section (Buchanan, ’22), but a day or two later such anesthetics in the same concentrations bring about no increase (Buchanan, unpub- lished). ; Concerning the experiments on P. lata, it may be noted, first, that the two factors are concerned in this species, and, second, that apparently nervous stimulation is less effective in decreasing and nervous inhibition in increasing head frequency than in P. doroto- cephala. Head frequency can be altered in both directions in P. lata, but so far as experiments have gone, it appears that general protoplasmic depressants, such as acids, are highly effective, while anesthetics in the stricter sense have little or no effect. These facts are in accord with observations on the living animal, which indicate that the nervous organization is considerably lower than in P. dorotocephala. PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 137 Another fact pointing in the same direction is that the original polarity can be more readily obliterated and biaxial heads produced by chemical agents than in P. dorotocephala. In certain concentra- tions of acids, for example, the frequency of biaxial heads is much increased. THE PHYSIOLOGICAL CONDITION OF PIECES FOLLOWING SECTION. Section of the body results in exposure of a cut surface which gradually contracts and within a few hours cell division and growth begin, giving rise to new embryonic tissue. It has been shown for P. dorotocephala that increase in rate of respiration and in susceptibility is slight or inappreciable in fourths or longer pieces, while in sixths and shorter pieces it is marked and increases as length of piece decreases, and also increases from anterior to posterior levels of the first zodid and decreases again in the poste- rior zooids (Child, ’14a; Robbins and Child, ’20; Buchanan, ’22; Hyman, ’22). Since these changes show definite relations to the polar gradient and are factors in determining head frequency (Child, 14d, ’16; Buchanan, ’22), it is of interest to determine whether similar changes occur and influence head frequency in le Manor Changes in Susceptibility Following Section—tThe data are most readily presented as graphs, plotted by the method used in earlier work (Child, ’15a, p. 81). This method is briefly as follows: Five stages in the progress of disintegration in KNC or some other agent from intact animals or pieces to completely disintegrated are more or less arbitrarily distinguished and these are given, respec- tively, the numerical values, 40, 30, 20, 10, 0. In determinations of susceptibility a certain number—e.g., ten—of animals or pieces is placed in the solution used and the number or individuals or Pieces in each stage is recorded at hourly or half-hourly intervals. The ordinate of the susceptibility curve for any time is the sum of the products obtained by multiplying the number for each stage by the numerical value of that stage, divided by the number of ani- mals or pieces in the lot. These ordinates are plotted against times in hours as abscisse. In the experiments on pieces the susceptibility of the fourths, sixths, and eighths was determined separately for each level of the 138 P. B. SIVICKIS. body; but since the pieces from different levels showed no great differences in susceptibility, the data for all the different levels were brought together in a single curve for each length of piece. Consequently each curve of pieces in the graph, Fig. 24, represents pieces from all levels—i.e., the whole body except the head cut into fourths, sixths, or eighths. KNC m/1,000 was used as agent. it Cae: Zao b a ee 30 20 0 + Hours CRC G Pe Mate fh 6 a Fic. 24. Graph showing changes in susceptibility following section. Each curve except aa and bb represents all pieces from 50 animals, aa, 50 intact animals, bb, 20 headless animals. Further explanation in text. Fig. 24 shows that susceptibility is greatly increased by section. Uninjured, full-grown animals (aa) show the lowest susceptibility . of all. Removal of the head increases susceptibility (bb). In fourths (cc), sixths (dd), and eighths (ee), immediately after section, susceptibility is greatly increased, and it will be noted that it is highest in eighths, somewhat lower in sixths, and lowest in fourths. During the first fifteen to twenty hours after section the sus- ceptibility of pieces decreases, then remains nearly stationary for PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 139 a day or two, and.then gradually rises as reconstitution progresses. Curve ff shows the susceptibility of fourths left in water fifteen to eighteen hours after section and then placed in KNC; curve gg, that of sixths under the same conditions; curve hh, that of eighths after forty hours in water. And, finally, curve i shows the sus- ceptibility of young growing animals 5 mm. long raised from pieces—t.e., approximately the susceptibility attained by fourths after reconstitution is completed—that of sixths and eighths being somewhat higher. Some part of this increase of susceptibility after section is un- doubtedly due to the presence of one (bb) or two (cc, dd, ee) cut surfaces. Other experiments for which curves have been plotted, but which are not shown here, demonstrate that pieces with oblique and therefore larger cut surfaces are more susceptible than pieces with transverse surfaces. | The experiments on susceptibility show one other point of im- portance which does not directly appear in Fig. 24 and which would require a number of graphs for full presentation. It was stated above that pieces of the same length from different levels show approximately the same susceptibility. In the section on | susceptibility gradients it was shown that susceptibility decreases from the anterior to the posterior end of the first zodid and in- creases again with approach to the level of the posterior zodéid. Even if the susceptibility of pieces from all levels were approxi- mately the same immediately after section, it would be evident that the increase in susceptibility must have been much greater in pieces from posterior than in those from anterior levels of the first zooid and from the posterior zooid. It follows from this fact that sus- ceptibility is not simply a matter of the presence of cut surfaces, but depends in part upon level of the body from which pieces are taken. Comparison of these changes in susceptibility following section with those in P. dorotocephala* brings to light some interesting physiological differences between the two species. In the first place, removal of the head does not appreciably increase suscepti- bility in P. dorotocephala. In fourths it is increased only slightly, 4See Child, ’t4a. His results have been repeatedly confirmed by myself and by other students in laboratory experiments, 140 P. B. SIVICKIS. in sixths somewhat more, and in eighths still more, but in all cases much less than in P. lata. Moreover, during the first twenty-four hours after section it usually decreases almost or quite to the same level as that of whole animals. In short, the presence of one or even two cut surfaces has in itself little or no effect in increasing susceptibility in P. dorotocephala. Differences in body level are far more important, particularly in the shorter pieces. Even in sixths or less the increase is slight in anterior pieces, becomes greater toward the posterior end of the first zooid, and is again slight in the posterior zooid. | Evidently the increase in susceptibility in relation to cut surfaces and the differential increase at different levels of the body depend, at least in part, on different factors. The former, which seems to be essentially a cellular wound reaction, followed by cell division and growth, is the more conspicuous feature in P. lata. The dif- ferential increase, on the other hand, is more conspicuous in P. dorotocephala, but is present also in P. lata, and appears to be a stimulation of the piece as a whole. Various facts, such as its short duration and its inhibition by anesthetics (Buchanan, ’22), indicate that it is nervous in character. Moreover, it is of interest to note that the differential increase varies inversely as suscepti- bility at different levels of whole animals and inversely as head frequency at different levels, while the increase in relation to cut surfaces varies directly with area of cut surface in relation to size of piece. Changes in Rate of CO, Production Following Section.—Colori- metric estimations of CO, production confirm the data on sus- ceptibility as indicative of changes in rate of respiration. In each of these experiments five 12-13-mm. headless animals entire were compared with five headless animals cut into eighths. The start- ing point was pH 7.9, and the pH was recorded at regular inter- vals, and the time required to reach pH 7.3 was also determined. After the experiment both lots were weighed, and in all cases the weight of the pieces was less by some 20 to 50 per cent. than the weight of headless animals, because some loss of intestinal con- tents, fluids, or even cells occurs when pieces are cut. In all cases, however, the rate of decrease of pH of the pieces was equal to or se PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. I41 higher than that of the entire headless animals. In Table III. the length of time required for change from pH 7.9 to pH 7.3 is given in hours and minutes for ten milligrams of headless animals and pieces, as calculated from the actual weights and times. This method of presentation is open to certain objections, but has the advantage of brevity and clearness. Of the nine experiments in Table III. five were with headless animals and eighths immediately after section, four with headless and eighths twenty-four hours (Arce ii Timez in Hours anD Minutes REQUIRED FOR CHANGE FROM PH 7.9 TO 7.3 CALCULATED FOR I0 MG, FROM AcTUAL WEIGHT AND TIME. Headless. Highths. Immediately after Section. 24 Hours a. s. MOSS Otay sre haici acto csv sh sieca ave 4:10 SEA Qh ga ktatncl cheesy severe <.als.as 4:57 ORS Olneee eis aie eicvouss ec aia e's 4:10 SDH Olepavey e aetekey elas es sy alls: ed ée 3:42 TeONA Oeweege sions aes cie ts Soa == 9:20 OR 2 Rea ue aretey aie fereiel isco: a8 — 5 OWA O Pare ME vohiccre ee “ele iss 8 4:53 = Sia Olena src lers wees behossdasiece 6:34 === Tis OMe anens ‘odometer == 4:44 IVC ATI MOA Gapeeietoserers cts, « ced nievs 4:51 5:51 after section. In every case the time is much less for the pieces, but is somewhat greater after twenty-four hours than immediately after section. These data confirm the susceptibility data in show- ing that rate of respiration in eighths immediately after section is greatly increased over that of headless controls, and that after twenty-four hours it is somewhat lower, but still far above that of the controls. Changes in Oxygen Consumption Following Section Table IV. gives the results of determinations of oxygen consumption made by Dr. Hyman. In the first experiment wholes, headless, and pieces are compared, in the others only headless and pieces, the headless animals being used because they show less motor activity than wholes. The pieces include both sixths and eighths from all levels posterior to the head. Table IV. shows that oxygen con- sumption is greatly increased in pieces immediately after section, as compared with headless animals, and in the first experiment in headless animals as compared with wholes. Twenty-four hours after section the oxygen consumption is lower in some cases, higher 142 P. B. SIVICKIS. in others, than immediately after section, and increase is more frequent in the pieces than in headless animals. Pieces of P. dorotocephala also show a similar increase in oxygen consumption AAweis DW. Oxycrn Consumption oF P. lata: WHotrE AnimMats, HEADLESS, AND SIXTH AND Er1GcHTH PiEcES IMMEDIATELY AND TWENTY-FouR HouRS AFTER SEcTION. CALCULATED IN CuBIc CENTIMETERS OF OXYGEN CONSUMED PER GRAM IN Four Hours. CALCULATIONS AT! 20° C, Headless. | Pieces. Experiment. Wholes, Immediate. 24 Hours. Immediate. 24 Hours. Tasheuciesustiels I.31 I.40 1.30 1.93 1.68 Pita 1.78 2A Eten emi Ne —— I.06 .99 I.44 Lost. Bie Stuy eters «a = I.34 Teall 1.60 1.83 Miaua. euch et isee == | 1.26 Wedel I.51 Walt I AEC eeore = I.00 I.20 I.59 I.79 ae eis 2.18 in some cases (Hyman, ’22). In this respect the data on oxygen consumption apparently disagree in part with those on CO, pro- duction and susceptibility. The reasons for this apparent dis- agreement are not as yet certainly known, but it seems probable that nutritive condition and growth at the cut surfaces which is appreciable within twenty-four hours are the factors chiefly con- cerned. . Time oF Heap DETERMINATION. It has been shown that in P. dorotocephala the head frequency characteristic of a certain length of piece at a certain level is deter- mined within a few hours after section to such an extent that it is only slightly or not at all altered by decreasing the length of piece after that time (Child, ’14d). The experiment to test this point gives essentially the same results in P. lata, except that the differ- ences in head frequency in long and short pieces are less than in P. dorotocephala. Fig. 25 shows the pieces used in the experi- ment. A large number of long pieces are cut with anterior ends PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 143 at level aa (Fig. 25), and at intervals the anterior ends of a certain number of them—e.g., fifty—are cut off as short pieces at the level bb and their head frequencies recorded after reconstitution. Omit- Fic. 25. Outline indicating levels of section, aa and bb, in experiment on head determination. ting tables and graphs, the most important results obtained are as follows: In the short pieces isolated at once in the usual manner the chief head frequencies are: normal, 40 per cent.; teratophthal- mic, 17 per cent.; acephalic, 30 per cent. In short pieces cut from anterior ends of long pieces after five hours head frequencies are: normal, 80 per cent.; teratophthalmic, 6 per cent.; acephalic, 2 per cent. In short pieces cut in the same way after twenty-three hours 100 per cent. normal heads appeared. This experiment shows that under ordinary conditions factors determining head frequency are to some extent effective during the first few hours after section, for during this period the altera- tion of head frequency characteristic of long pieces by great. de- crease in length of piece becomes progressively less. This, of course, does not mean that head frequency can not be altered in other ways after this time; it merely means that conditions deter- I44 P. B. SIVICKIS. mining head frequency are to some extent established during this time. Since this is the period of greatest stimulation of pieces, since that stimulation is greater in shorter than in longer pieces and in posterior than in anterior pieces of an individual or zooid, and since decrease of that stimulation during the first few hours after section increases head frequency, it appears highly probable that the differential stimulation of pieces after section is a factor in determining that under ordinary conditions head frequency is lower in shorter than in longer and in posterior than in anterior pieces of individual or zooid. RATE OF GROWTH AND DIFFERENTIATION AND REGIONAL DIFFER- ENTIALS IN DEVELOPMENT AT ANTERIOR CUT SURFACES. Only a brief statement of general results along these lines is given at the present time. The growth reaction at anterior cut surfaces is much more rapid in P. lata than in P. dorotocephala. In the former the strong contraction following section has very largely disappeared and a distinct outgrowth of new tissue is present over the whole surface twenty-four hours after section (Fig. 26), while in the latter the cut surface is still strongly con- tracted and there is less than half as much new tissue (Fig. 27). On posterior surfaces the differences between the two species are less marked (Figs. 26, 27). Two days after section the differ- oe a be a Fics. 26, 27. New tissue twenty-four hours after section: Fig. 26, P. iata; Fig. 27, P. dorotocephala. ences between the species remain about the same, but during the third day the rate of growth in P. dorotocephala becomes more rapid, as compared with that in P. lata, and on the fourth day the amounts of new tissue are apparently about the same. Evidently the initiation and acceleration of growth at the cut surface occurs — PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 145 earlier in P. Jata than in P. dorotocephala. ‘This difference prob- ably accounts, at least in part, for the fact that susceptibility and rate of respiration in pieces of P. Jata remain considerably higher after section than in whole animals, while in P. dorotocephala the decrease within twenty-four hours after section is greater, often to the level of whole animals. The rate of growth and of differentiation of the head, as deter- mined by the time at which the eyes become visible, is approxi- mately the same in anterior regions and in the posterior zooid in both species, and in both it is more rapid in these regions than at posterior levels of the first zooid. It also decreases and the dif- ferences in rate at different levels increase with decreasing length of piece. In other words, curves of rate of differentiation of head plotted from repeated observations of developing pieces re- semble in their relations to body level and length of piece the head-frequency curves. Comparison of the two species shows, however, that in the shorter pieces and the more posterior levels of the first zooid the rate of differentiation is slightly lower in P. lata than in P. dorotocephala. That is to say, in P. lata the head frequency in such pieces is higher, but the rate of differenti- ation of the head is lower than in P. dorotocephala. As regards the portion of body posterior to the head which is formed by new tissue, there is little difference at different levels of section in longer pieces, but with decreasing length of piece this differential appears. In eighths, for example, from levels near the anterior end of the animal and near or in the posterior zooid, the eyes develop at or slightly anterior to the boundary between new and old tissue (Figs. 28, 29), while in pieces from the poste- 30 28 29 Fics. 28-30. Regions of body formed by anterior new tissue at different levels in eighths of P. lata: Fig. 28, anterior region of first zodid; Fig. 29, posterior zooid; Fig. 30, posterior region of first zooid. 146 P. By SLVICKIS, rior region of the first zooid—+.e., about the level of the pharynx— a considerable portion of the body posterior to the eyes develops from new tissue (Fig. 30). A similar regional differential ap- pears in shorter pieces in P. dorotocephala and P. maculata, but the amount of difference differs somewhat in the different species. In P. foremannii and other species which possess no posterior zooid the regional differential continues to change in the same direction to the posterior end of the body—+.e., the more posterior the piece, the longer the portion of the body formed from new tissue at the anterior end (Morgan, ’or). This regional differential is evidently determined by a complex of factors—e.g., rate of growth of new tissue, size of new head, degree of inhibition in its development by stimulation of the piece, rapidity of reorganization of old parts, and perhaps others. The chief point of interest at present, however, is the fact that this regional differential shows the same relation to body level and length of piece as does head frequency. Its relation to the polar axial gradient is therefore evident. Moreover, experiments show that this differential can be altered and controlled by the same factors by which head frequency is altered and controlled. DISCUSSION. Reconstitution in Relation to Body Level, Length of Piece, and Physiological Age.—It is evident that the axial susceptibility gra- dient is an indicator of fundamental physiological differences along the axis and many facts indicate that such differences are pri- marily quantitative rather than qualitative. Head frequency, dif- ferential increase in susceptibility following section, rate of growth of new tissue at the anterior end, and the portion of the body posterior to the eyes which is formed by new tissue all show a gradation according to body level and therefore a definite relation to the susceptibility gradient. This is true for P.. dorotocephala as well as for P. lata. That the physiological factors which deter- mine these graded differences in reaction to section and in the processes of reconstitution are fundamentally quantitative, not qualitative, is indicated by the fact that there is no evidence of fixity or specificity in their relation to body level. All the features of reconstitution characteristic of a given body level under normal PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 147 or standard external conditions can be altered to those character- istic of other levels by changes in external conditions which are primarily non-specific and quantitative in physiological effect. This has been shown for P. dorotocephala in many ways and is also true for P. lata, though only a part of the evidence appears in the present paper. It has also been shown for P. dorotocephala that the susceptibility gradient is an indicator of a corresponding gradient in rate of respiration (Robbins and Child, ’20; Hyman, 22), and the data on the changes in susceptibility and rate of respiration in pieces following section leave no doubt that a similar relation between susceptibility and respiratory rate exists in P. lata. If this is true, the inference is justified that the reconstitu- tional differences at different levels are in some way associated with the differences in rate of metabolic reactions, as indicated by rate of respiration. The relations of reconstitutional processes to length of piece also appear to be non-specific and quantitative in character, for they, too, can be altered by the quantitative action of external factors. And, finally, the relation of head frequency to physiological age is apparently non-specific and quantitative and can be altered ex- perimentally by changes in condition which affect primarily rate, rather than kind of metabolic reactions. Nowhere do we find any evidence for the existence of specific formative substances. Given the specific protoplasm of a plana- rian species, the differences in the reconstitutional processes and results are apparently dependent primarily upon quantitative dy- namic differences rather than upon specific qualitative factors. Physiological Analysis of Head Frequency.—It has been shown that the head forms differing from the normal which appear in the reconstitution of P. dorotocephala represent various degrees of differential inhibition of head development, and that they can all be produced by chemical and physical agents, as well as by physi- ological factors (Child, ’16, ’20a, ’21; Behre, ’18; Buchanan, ’22). All the experimental evidence supports the conclusion that two antagonistic factors are concerned in the reconstitution of a head— the one positive or determining, the other negative or inhibitory— and that head frequency in any particular case depends on the relation between these two factors (Child, ’14a, ’14d, ’16, ’20a, ’21; 148 P. B. SIVICKIS. Behre, 18; Buchanan, ’22). The evidence indicates further that the positive determining factor is the rate of activity of the cells at or near the cut surface (Fig. 31, +) which react to section by ——————— Fic. 31. Diagrammatic outline of piece after section: +, region directly con- cerned in formation of head; y, region of correlative inhibitory effect on head development. dedifferentiation, division, and growth, and are directly concerned in head formation. The inhibiting factor, on the other hand, is apparently the correlative influence on x of other parts of the piece (Fig. 31, y) which tends to retard or inhibit the dedifferenti- ation and growth of the x cells. More or less excitation of the region y occurs temporarily after section, probably largely because of the injury to the nerve cords, and experiments have shown, first, that head frequency decreases as the degree of this excitation increases—e.g., at the more posterior levels of the first zodid and in shorter pieces—and, second, that inhibition of this excitation increases head frequency. It has been shown further that the differential susceptibility of the regions + and y and the different degrees of excitation of y at different levels of the body provide a physiological basis for altering head frequency experimentally in either direction with the same concentration of a single chemical agent (Child, ’16) and with many different agents and conditions (Child, ’20a; Behre, 718; Buchanan, ’22). In short, the facts indicate that head frequency varies directly with rate of metabolism in + and inversely with rate in y. This relation has been stated in the following brief form: tate + head frequency = eae: This formula is perhaps not complete, but serves provisionally to indicate the opposite relation of the two factors to head frequency. If it indicates the relation correctly, it follows that head formation in reconstitution really takes place in spite of the rest of the piece. ee ee eee = as PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. 149 In other words, in so far as the wx cells become independent of y, they dedifferentiate and begin the development of a new individ- ual, the head arising first as in embryonic development. The vari- ous differentially inhibited head forms result from different de- grees of inhibition of 4 by y or by some external factor. All the experimental evidence at hand indicates that this inter- pretation holds for P. lata as well as for P. dorotocephala, the chief difference being that in P. Jata the inhibiting action of y shows less increase with decrease in length of piece and at more posterior levels of the first zodid, and that head frequency is there- fore normally higher in the shorter pieces and at more posterior levels of the first zodid and is less increased by inhibition of y in P. lata than in P. dorotocephala. This difference between the species is what we should expect if the inhibiting action of y on head formation is nervous in character, as the facts lead us to believe. The differences in excito-motor behavior and in develop- ment of sense organs certainly indicate a lower degree of nervous organization in P. lata than in P. dorotocephala. Moreover, the differences in degree of excitation of y following section and in head frequency in relation to level of body and length of piece are less in P. lata than in P. dorotocephala, and this again suggests a lesser degree of specialization of different body levels in relation to the axial gradient. The fact that in P. lata head frequency is almost as high in young as in old animals, while in P. dorotocephala it is much lower in young than in old, also indicates that the region y is less effective in inhibiting head formation at x in P. lata. Rate y is higher in relation to rate x in young than in old animals because the tissues of young animals have in general a higher rate of metabolism than those of old, but this difference has much less effect on head fre- quency in P. lata than in P. dorotocephala. The more rapid growth reaction in P. lata also indicates that y is less effective in this species in inhibiting dedifferentiation and growth of +. The physiological analysis of reconstitution leads, in fact, to the same conclusion as observations on the behavior of the two species, viz., that P. lata is a more primitive, a less highly specialized form than P. dorotocephala. Tailless Forms and Biaxial Heads—Tailless forms develop 150 P. B. SIVICKIS. from very short pieces in which a single polarity, in these experi- ments the original polarity, is maintained. They arise when. the cells at the posterior cut surface are not sufficiently active in rela- tion to parts anterior to them to grow at the expense of the latter. Heads cut off immediately behind the eyes always remain tailless, and in general when the piece.is so short that the posterior cut surface is very close to the new head the development of a new posterior end is inhibited, because the rate of metabolism at other levels of the piece is so high that the cells at the posterior end can obtain but little nutrition. Biaxial forms arise in very short pieces of Planaria when a new physiological gradient arises in relation to the posterior cut sur- face. In the short piece conditions are most favorable for the origin of such new gradients because there is but little physiologi- cal difference between the two cut ends—z.e., these short pieces are nearly apolar because they are short, consequently each cut surface may become a dominant region and determine a polarity in the opposite direction to the other (Child, ’15b, pp. 98-100). The higher frequency of biaxial heads under ordinary conditions in P. lata than in P. dorotocephala suggests that polarity—1z.e., the longitudinal axial physiological gradient—is less stable in the for- mer, and this also is in accord with the conclusion that P. lata is a less specialized form than P. dorotocephala. SUMMARY. 1. Planaria lata possesses a short posterior zodid and undergoes fission. Fission does not interfere with sexual maturity, probably because the level of fission is far posterior to the genital pore. . 2. Susceptibility decreases from the anterior to the posterior -end of the first zodid and increases again with approach to the posterior zooid. 3. Head frequency varies in relation to level of body in the same way as susceptibility. The differences in head frequency at different levels increase as length of piece decreases, and head frequency is slightly lower in young than in old animals. 4. Isolation of pieces is followed immediately by a great increase in susceptibility, CO, production, and oxygen consumption, then a gradual: decrease occurs during 12-24 hours to a level still far PHYSIOLOGY OF RECONSTITUTION OF PLANARIA LATA. I51 above that of whole animals. Later a gradual increase coincides with the progress of reconstitution. 5. The increase in susceptibility after section in relation to body level and length of piece varies inversely as the susceptibility gradient of whole animals and the head frequency in reconsti- tution. 6. The factors determining whether a piece shall or shall not develop a head become to some extent effective within a few hours after section. 7. The experimental data all support the conclusion that in P. lata, as in P. dorotocephala, two factors are concerned in deter- mining head frequency: one the rate of metabolism in the cells at the cut surface, being positive or determining; the other the rate of metabolism in other regions of the piece, being negative or inhibitory. Head frequency in any particular case is determined by the relation between these two factors. Apparently the in- hibitory factor is less effective in the shorter pieces and at more posterior levels in P. lata than in P. dorotocephala. 8. The physiological analysis of reconstitution agrees with ob- servations on behavior in indicating that P. lata is a less highly specialized form than P. dorotocephala. REFERENCES. Behre, Elinor H. 18 An Experimental Study of Acclimation to Temperature in Planaria dorotocephala. Biot. Buty., XXXV. Buchanan, J. W. "22 The Control of Head Formation in Planaria by Means of Anesthetics. Jour. Exp.-Zool., XXXVI. Child, C. M. "10 ©6Physiological Isolation of Parts and Fission in Planaria. Arch. f. Ent- wicklungsmech., XXX., II. Teil. 11a Experimental Control of Morphogenesis in the Regulation of Planaria. Biot. BuLtu., XX. ’r11b Studies on the Dynamics of Morphogenesis and Inheritance in Ex- perimental Reproduction. I. Jour. Exp. Zodl., X. "r1c ©6Studies, etce., III]. Jour. Exp. Zodl., XI. ’13a Studies, etc., V.- Jour. Exp. Zo6l., XIV. "13b ©Studies, etc., VI. Arch. f. Entwickelungsmech., XXXVII. "14 Studies, etc., VII. Jour. Exp. Zo6l., XVI. "tab Asexual Breeding and the Prevention of Senescence in Planaria velata. Brot. Buty., XXVI. ’14c «Starvation, Rejuvenescence and Acclimation in Planaria. 152 P. B. SIVICKIS. ’14d Studies, etc., VIII. Jour. Exp. Zoél., XVII. 14e Starvation, Rejuvenescence and Acclimation in Planaria dorotocephala. Arch. f. Entwickelungsmech., XXXVIII. "15a Senescence and Rejuvenescence. Chicago. ’15b Individuality in Organisms. Chicago. 716 Studies, etc., IX. Jour. Exp. Zodl., XX1. Iga A Comparative Study of Carbon Dioxide Production in Planaria. Amer. Jour. Physiol., XLVIII. 19b Susceptibility to Lack of Oxygen During Starvation in Planaria. Amer. Jour. Physiol., XLIX. ’20a Studies, etc., X. Jour. Exp. Zoél., XXX. 20b Some Considerations Concerning the Nature and Origin of Physio- logical Gradients. Brot. Buty. XXXIX. *21 Studies, ete., XI. Jour. Exp. Zoél., XXXIII. Child, C. M., and McKie, E. V. M. "11 The Central Nervous System in Teratophthalmic and Teratomorphic . Forms of Planaria dorotocephala. Buior. Buru., XXII. Curtis, W. C. 702 :«~The Life History, the Normal Fission and the Reproductive Organs of Planaria maculata. Proc. Boston Soc. Nat, Hist., XXX. Hyman, L. H. ’t9a ©Physiological Studies on Planaria. 1. Amer. Jour. Physiol., XLIX. ’19b Physiological Studies on Planaria. II. Amer. Jour. Physiol., L. ’19¢_ ~«=9Physiological Studies on Planaria. III. Brot. Bury., XXXVII. "20 Physiological Studies on Planaria. IV. Amer. Jour. Physiol., LIII. ’22 Physiological Studies on Planaria. V. Jour. Exp. Zodl. Morgan, T. H. ?0r Growth and Regeneration in Planaria lugubris. Arch. f. Entwickelungs- mech., XIII. Robbins, H. S., and Child, C. M. ’20 Carbon Dioxide Production in Relation to Regeneration in Planaria dorotocephala. Biot. BuLy., XXXVIII. ' es eT eS ee ee ee | BIOLOGICAL BULLETIN Marine Biological Laboratory eae : | WOODS HOLE, MASS. } Vou. XLIV PAE RIE 102g ne Os ‘CONTENTS | BY McNam, GEoRGE T. Motor Reactions of the Fresh-Water. , Sponee, ; x 3 Ephydatia fluviatilis ALLEE, W. Cor Studies in Marine Ecology. 1, The Distribution NER of Common Littoral Invertebrates of the Woods Hole Region. Bm UR ORE OR cae uae ca Aa £167 fe; ‘LAMBERT, We Ves .. Food and Parthenogenetic Reproduction as Relat- ay ‘RicE, W. S. AND ed to the Constitutional Vigor of Hydatina senta 192 WALKER, H. C- A. Ms : PuBLIsHED MONTHLY BY THE MARINE BIOLOGICAL LABORATORY PRINTED AND ISSUED BY THE NEW ERA PRINTING COMPANY, Inc. LANCASTER, PA. a . ‘ | AGENT FoR, GREAT Britain moo _.. WHELDON & WESLEY, Liurep oe 2,3 and 4 Arthur Street, New Oxford Street, London, W. P. 2 — y Single Numbers, 75 Cents. Per Volume (6 numbers), $3.00 ak ‘Entered October 10, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16. 1894. Editorial Staff. E.G. Conxiin—Princeton University. _ GrorcE T. Moore— Zhe Missouri Botanic pare T. H. Morcan— Columbia University. - W. M. WHEELER— Harvard University. E. B. Witson— Columbia University. Managing Lditor Frank R. Lituiz— The University of Chicago. All communications and manuscripts should be sent to the Man- aging Editor, the University of Chicago, Sept. 15th to June 15th, or Woods Hole, Mass., June 15th to Sept. 15th. Subscriptions. and other matter should be addressed to the Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa. Vol. XLIV A pril, 1023. No. 4. BIOLOGICAL BULLETIN MOTOR REACTIONS OF THE FRESH-WATER SPONGE, EPHYDATIA FLUVIATILIS} GEORGE T. McNAIR, HULL ZOOLOGICAL LABORATORIES, THE UNIVERSITY OF CHICAGO. Parker’s work on Stylotella heliophila (10) has suggested the possibilities of similar experiments with fresh-water sponges and a comparison of the reactions of the two types. Aside from Parker’s monograph practically nothing has been published on the motor reactions of sponges. Dr. R. E. Grant (1825—’26) published a series of papers on the “Structure and Functions of Sponges,’’ which are of historical interest only. He quotes an earlier paper (Ellis and Knight, Transactions of the Royal Society of London, 1765), in which these men stated that they had seen the orifices on the surfaces of sponges contract and dilate themselves. Grant studied a large number of salt-water species, both in their natural habitat and in the laboratory. He was the first to notice and accurately describe the currents of water coming from the fecal orifices (oscula) and spoke of them as mysterious currents. because he could not learn what caused them. He irritated the orifices and other tissues with corrosive acids, red hot wires, etc., but under no circumstances did he see the osculum closed and erroneously concluded that there was a complete lack of irritability and contractility, attributing the statements of Ellis and Knight concerning the “systole and diastole of the fecal orifices” to ‘‘some optical deception a little assisted by the imagination.” Parker made two types of experiments on Stylofella, noting 1The problem was suggested to me by Dr. W. J. Crozier. It is with great pleasure that I acknowledge my indebtedness to him and also to Professors C. M. Child and W. C, Allee for their helpful suggestions and criticisms. 153 II 154 GEORGE T. MCNAIR. the responses of the oscula, ostia, and choanocytes first to mechanical and then to chemical stimuli. He studied the reactions of the sponge to flowing water, injuries, and cuts. Likewise he studied the effects which ether, chloroform, cocaine, strychnine, and atropine produced upon the sponge and compared his results with the known effects of these drugs on smooth muscle and on ciliated tissue. , Ephydatia fluviatilis (L.), the fresh-water sponge with which the experiments described in this paper were made, differs so greatly in structure from Stylotella that it was impossible to duplicate many of Parker’s experiments. The responses of the oscula were watched to determine the effects of various mechan- ical stimuli such as injuries, electrical stimulation, and changes in temperature. The purpose was to determine the effects of such stimuli and the rate of possible transmission of these effects from one part of the body to another. All the sponges used in the following experiments were collected in small ponds along the railroad tracks in the neighborhood of Hammond and Buffington, Indiana. They were carried to the laboratory in fruit jars and then transferred to large jars of slowly running water. The ponds from which the material was collected! were shallow, often almost filled with bulrushes. The bottoms were covered with mud and cinders. The sponges were, for the most part, growing on the under side of old railroad ties floating in the water and were more or less cushion-like in form and without much branching. Some, however, were found growing on the submerged parts of plants or on leaves, but here their form was more spreading, following that of the body to which they were attached. The body of the sponge is dotted all over with very small dermal pores or ostia. The oscula are relatively prominent little chimneys of dermal membrane, standing our from almost any point on the suface. Bundles of smooth, pointed, and almost straight silicious spicules make up the skeleton. The internal canals of the body meander irregularily between these bundles of spicules. The dermal membrane, which completely envelops 1 See Shelford, ‘‘Animal Communities in Temperate America,’’ Chapter VIII. es Se MOTOR REACTIONS OF THE FRESH-WATER SPONGE. 155 the body, is held up.in tent-like elevations by the protruding ends of the spicules. The oscular chimneys seem to be kept open by the pressure of the currents of water passing out through them. They are continuously changing their form. At times one will be long and slender; again it will be shorter and much thicker or even dome-like, the base being as much as three or four times the height. In instances of the first type the mouth of the chimney will probably be open almost to the width of the diameter, although these mouths are frequently seen to be nearly closed. In the type last mentioned they are usually small. SQN OD AD AVN ASS OMAN LMA MAMAS ee . AL SOOO YALA ACRE, Spree ee a D E F Fic. t. Various forms in which the same oscular chimney was observed. These variations in shape lead to the belief that the entire chimney contains sphincter-like bands. If all such bands were about equally contracted the result would be a long slender chimney, Fig. 1, A and B. If only those at the end contracted and the others relaxed, the result would be the dome-like shape, Fig. 1, C and D. Several instances of a contraction of the terminal sphincter and also of the chimney near its base, with a relaxation between, were noticed. The result was a sort of globe-like chimney attached to the sponge body by a narrow neck, Fig. 1, H. The chimneys also seem capable of a decided shortening without much change in diameter. This may be 156 GEORGE T. MCNAIR. brought about by a contraction of longitudinal nner but it could not be definitely demonstrated.! EFFECTS OF CURRENTS OF WATER ON THE OSCULUM. Potts (’87) makes the statement that he found £. fluviatilis in both quiet and running water. It seems to grow best in ponds of quiet water, but very fine specimens are found on the under side of rocks in rapids or at the base of water falls where the river current is the swiftest. To get an idea of the effects of water currents, a few sponges were placed in a flat dish containing about a quart of water and left over night. The next morning the oscula were still expanded, and, by watching the movements of tiny particles of suspended material in the water, steady currents could be seen flowing through them. This test was repeated with the same results. It was noticed, however, that if the sponges were left in such a dish of unchanged water much longer than a day they gradually became less active and in many cases soon died. To determine the effect of a swift current, a glass tube was connected with the water tank and placed in one side of a dish containing the sponges. The water was turned on at full force and left running for about four hours. After transferring the sponges to a watch glass for observation (without taking them from the water) it was found that the entire chimney had con- tracted until it was almost flat, but that the mouth of the osculum was wide open and currents of water were coming out. The following illustration shows the extent of the contraction of MOM NTA WOES SS > the chimney. Fic. 2. A chimney before and after a strong current of water had been applied, showing the extent of the contraction. 1See, E. A. Minchin, Lankester’s ‘‘Treatise on Zodlogy,”’ p. 44. > a { MOTOR REACTIONS OF THE FRESH-WATER SPONGE. 157 Within thirty minutes after taking the sponges out of the swift current the chimney had again extended to about half of its original length. The osculum was slightly contracted. The current had caused the sponges to roll and tumble over the bottom of the dish to some extent. To find out if this had been the cause of the shortening of the chimneys, the sponges were placed in watch glasses which, in turn, were placed within the larger dish. When the water was turned on, the force of the current against the curvature of the watch glass held the sponges in place. This time a swift current flowed over the sponges but they did not roll about. Within twenty minutes the chim- neys had flattened out as before. The sponges were now placed in a large jar in which a slow current of water was kept going. After being there for a few hours the chimneys were found to be well extended and with strong currents coming out through them. Neither the absence of currents nor the presence of strong currents causes the osculum to close although the latter does cause a general shortening of the oscular chimney. In small quantities of water, however, the sponges soon became less active and in time died; the smaller the quantity of water the sooner the changes could be noticed. This effect could not have been due to the absence of the mechanical stimulus of flowing water, because if placed in a large tank of quiet water the sponges will live for a long time. Nor could it have been caused by lack of oxygen, because the dishes which were used were large and shallow so that the water was well exposed to the air. It must have been produced then by self-poisoning from the prod- ucts of metabolism that accumulated’in the small quantity of water. The greater general vitality of the sponges when placed in slow running water or in large tanks of quiet water is accounted for by the fact that these products were removed as they were thrown off by the sponges or were diffused through the large volume of water. These results are entirely different from those obtained by Parker from somewhat similar tests with Stylotella. When he transferred Stylotella from the natural habitat to tanks of quiet sea water, the oscula invariably closed within ten minutes and 158 GEORGE T. MCNAIR. did not reopen until they were placed in running water. He was able to prove that the mechanical stimulus of the currents of water on the outlet tip of the finger where the osculum was located, caused it to open. The oscular closure seems to be a protective adaptation of Stylotella, which is a shallow salt-water species frequently left exposed to the air when the tide is out. Since the fresh-water sponge is never exposed to such tidal conditions, one would not expect to get the same reactions. BRUSHING. Rubbing the sides of the chimney with a needle, prodding it gently, or even inserting the needle into the mouth of the osculum and rubbing the inside of the chimney, caused no noticeable reaction; however, if the needle were carefully rubbed around the edge of the mouth of the osculum, an immediate contraction of the orifice would follow and it would continue to contract for about three minutes. Following the contraction of the sphincter at the orifice a wave of contraction would travel down the chimney although the amount of contraction would be less than at the orifice. On a chimney of 1.5 mm. length it would require about six seconds for this wave to run from the tip to the base of the osculum. In the next three or four minutes it would again expand to its original size. The edge of the mouth of the osculum, therefore, seems to be more sensitive than the rest of the chimney. Concerning the reactions of Stylotella, Parker states that stroking an open osculum with a bristle or brush does not cause it to close, but if it is closed, this treatment may at rare intervals induce it to open. SHARP BLow. Several times while getting the electrodes into position to work with the inductorium one of the wires of the electrode would catch on a spicule and, springing loose, strike an oscular chimney. Each time the entire chimney would shrivel up immediately in a more or less collapsed condition, but within twenty or thirty minutes it would be open and functioning as before. CUTTING. Cutting into the body seemed to have no effect in the way of induced:movements. With a small scalpel an incision was made eee eT ee a ag thar M4 MOTOR REACTIONS OF THE FRESH-WATER SPONGE. 159 into the body of the sponge 3 mm. from the base of the chimney without causing any change in the rate of the currents or in the size of the chimney. Another cut right at the base of the chimney caused it to close. It did not reopen for two days. This cut was so near the chimney, however, that the closing was probably due to cutting into the chimney cavity and when this had healed over the chimney again expanded. The result of such injuries — differs from the effect of similar ones on Stylotella, where a cut from 3 mm. to 5 mm. from the osculum was found to cause closure within nine minutes. PIN STICKING. Sticking a needle into the flesh of the sponge at a distance of 2 mm. from the osculum caused no noticeable effect on the osculum. The same type of injury 5 mm. from the osculum of Stylotella resulted in the osculum closing within ten minutes and remaining closed for several hours. EXPOSURE TO THE AIR. A sponge was lifted from the water and held in the air for three minutes. A large chimney was noticeable before, but — could not be found after placing the sponge back in the water; neither could it be found at any later time. THERMAL RESPONSES. The average temperature of the water in the tanks in which the sponges were kept was about 27° C. Eight of the sponges which had well developed oscular chimneys and from which strong currents of water were flowing were placed in dishes in a refrigerator which had a uniform temperature of 5° C. and left until the water had been cooled to 7° C. They were examined at regular intervals. At 16° there was no noticeable change in any of the sponges. At 11° all had a shriveled appearance, some of the oscula were open slightly, and very slow currents coming from them were barely visible. At 10° no oscula could be found, and at 7° the dermal membrane had completely shrunk in over the body of all the sponges and no sign of activity could be detected. The dishes were transferred to an incubator and 160 GEORGE T. MCNAIR. observations were made until the temperature of the water had reached 40° C. At 24° the dermal membranes had taken on their usual appearance, the oscula were somewhat expanded and currents were issuing from them. At 34° the chimneys were more extended and stronger currents were coming from them than at any previous time. At 37° they were still expanded but were slightly flabby and the currents were weaker. At 40° the oscula had all disappeared, the dermal membranes were com- pletely shriveled. None of the sponges recovered after being raised to the temperature of 40°. Comparing these results with those obtained by Parker, it will be seen that the salt-water sponge is apparently less affected by temperature conditions than is the fresh-water species. With Stylotella at between 9° and 10° the oscula and ostia remained open and the only effect seemed to be that the currents became slow. At 40° there was a slight constriction of the osculum and the currents gradually became slower and stopped. At 45° there was a flabby contraction of the oscula and the currents ceased abruptly. They did not recover after being raised to the temperature of 45°. : Ere@na: Neither full sunlight nor complete darkness seemed to have any effect on E. fluviatilis. One large, healthy specimen was placed in a dark cupboard and left there four days. At the end of this time the green color due to contained alge had faded, but the activity of the sponge in producing currents had not diminished in any way. ELECTRICAL STIMULATION. By using an inductorium with electrodes of fine platinum wires, the effect of weak Faradic stimulation could be observed. The electrodes were applied to three.different places: first, on the side of the chimney; second, at the base of the chimney, one wire on each side; third, at points on the flesh immediately back of the chimney at distances of from 3 mm. to 5 mm. from it. A weak stimulus when applied to the tip of the chimney caused a contraction of the oscular sphincter and a gradual LEAR LEN OES MOTOR .REACTIONS OF THE FRESH-WATER SPONGE. 161 wave of contraction down the entire length of the chimney. This wave traveled down at the rate of about 1.5 mm. in four seconds. A strong stimulus applied to the chimney caused a complete relaxation, as though paralyzed, and from which it did not recover. STIMULATION AT THE SIDES OF THE CHIMNEY. Choosing chimneys which were long and well extended the electrodes were applied at one side of the chimney, Fig. 3, A. When a weak stimulus was used, the response required about one minute before it appeared. A mild stimulus resulted in an immediate response. First, the chimney would bend toward the electrodes, the bending occurring at the points where the wires touched the chimney, Fig. 3, B. This bending always occurred within thirty seconds from the time the stimulus was started. At the same time, but continuing longer, the entire chimney would shrink slightly. This slow shrinking continued for from one to five minutes. If the wires were applied to the opposite side (Fig. 3, C) while the chimney was still bent as in (B), within one minute the end of the chimney would swing over in that direction, Fig. 3, D. Within thirty minutes the chimneys would straighten and expand to their former size and shape. bal D Fic. 3. The arrows indicate the points on the side of the chimney where the electrodes were applied. STIMULATION AT BASE OF OSCULUM. When a very weak stimulus was given for fifteen seconds with an electrode at each side of the base of the chimney, a very gradual longitudinal shrinking could be observed. The time required for the contraction to travel from the base to the tip was ten seconds, a rate of transmission of 0.17 mm. per second. 162 GEORGE T. MCNAIR. If the stimulus was a little stronger, an immediate contraction was followed by a slower general contraction, both lengthwise and in circumference, which would continue for about fifteen minutes, Fig. 4, 4 and B. The chimney would then begin]to expand and within thirty-five minutes would be almost as large as before. On one occasion the sphincter at the base contracted more quickly than the others, resulting in the form of chimney shown in Fig. 4, C. S A B C Fic. 4. The arrows indicate the points on each side of the base of the chimney where the electrodes were applied. STIMULATION ON THE FLESH. On several sponges the electrodes were applied immediately behind the osculum at distances of from 4 mm. to 2 mm. from it, and the flesh stimulated for periods varying from fifteen to sixty seconds. No effects could be seen on the chimneys. There was, however, some indication of an excitation of the choanocytes. The currents were stronger and more rapid for from ten to fifteen minutes after the stimulation. Although this was taken as an indication of greater activity on the part of the flagellate cells, it is possible that the stimulation caused the ostia to dilate or open wider and in this way allow a larger volume of water to pass into the sponge body. It does not seem, however, that this would account for the increased rapidity with which the currents poured out of the oscula. There was a noticeable difference in the responses of the oscula to Faradic stimulation depending on the strength of the electrical current used. From a weak stimulus,! a very gradual contraction was followed by a more rapid relaxation, while, if 1 By a weak stimulus is meant one which could barely be detected when the electrodes were applied to the tongue, while by a mild stimulus is meant one that could be felt when the electrodes were applied to the lips. MOTOR REACTIONS OF THE FRESH-WATER SPONGE. 163 the stimulus was mild, there would be an immediate contraction followed by a slower more general contraction. In this case the chimney would remain contracted for a few minutes and then expand much more slowly than it had contracted. In the second instance, all of the contractile cells probably were directly stimulated by the electrical current, while in the first case it is probable that only those where the electrodes were applied were directly stimulated, the gradual contraction being the result of transmission of the stimulus from one cell to the next. In one instance, following a very weak stimulation for thirty seconds, it was possible to keep an accurate measurement of the time required for contraction and relaxation of the sphincter at the mouth of the osculum. No check was made on the general contraction or relaxation of the chimney. That the contraction was slower than the relaxation is shown by the following graph. The temperature of the water was about 27° C. SN Se eee Sea en 2 [a] + [ala PEE nae oat a ce Ta es A SS a a Min. 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Fic. 5. Graph showing the rate of contraction and of relaxation of the sphincter at the mouth of an osculum following a weak Faradic stimulation. The abscissa represents the intervals of time in minutes and the ordinate the diameters of the opening in millimeters. TRANSMISSION OF STIMULI. It has already been pointed out that, apparently, very little transmission from one part of the body to another resulted from the application of the various types of stimuli used in these experiments. To determine if there was any transmission at all, 164 GEORGE T. MCNAIR. a sponge which had two large oscula at a distance of about 5 mm. apart was selected. A strong stimulus was applied with the electrodes at the base of one chimney. The result was an immediate jerk of the entire chimney followed by a gradual and complete relaxation of the entire chimney from which it did not recover. The other chimney was not affected. There is, however, a noticeable transmission of stimuli in the chimney itself and the rate of this transmission is more rapid from the tip to the base than from the base to the tip. This is shown in the experiments with electrical stimulation and by rubbing the chimney with a needle. Rubbing the edge of the mouth of the osculum with a needle started a downward wave of contraction which could be seen to travel at the rate of about 0.25 mm. per second. Rubbing at the base produced no effect. Weak Faradic stimulation at the tip caused a downward wave of contraction with a rate of about 0.35 mm. per second, while stimulation at the base produced a slight wave of contraction which traveled toward the tip at the rate of about 0.17 mm. per second. CONCLUSIONS. Ephydatia fluviatilis is a fresh-water sponge which lives in either quiet or flowing water. Swift currents cause a shortening of the oscular chimney but do not cause a constriction of the osculum. Placing the sponge in a small quantity of quiet water ‘results in a cessation of activities and death of the sponge in a short time, probably because of the metabolic products which accumulate in the water and poison the sponge. The edge of the mouth of the osculum seems to be more sensitive to stimulation that the remainder of the chimney, as shown by the effect of rubbing it in various places with a needle. Cutting the body of the sponge or sticking a needle into it seemed to have no effect except the local effect on the tissues which were injured. That there is no transmission of the effects to surrounding tissues, even for a distance of 2 mm. or 3 mm. from the injury, is shown by the continued functioning of the parts as indicated by the currents coming from the oscula and Ais me bi: tipo MOTOR REACTIONS OF THE FRESH-WATER SPONGE. 165 by the fact that the chimney did not contract or in any way change its shape. Subjecting the sponge to a low temperature does not have as disastrous an effect as increasing the temperature. Between 20° amd 30° C. seems to be approximately the temperature at which it thrives best. It is unable to live at a temperature of 40° C. The oscular chimneys being continuous with the dermal membranes and themselves being contractile, one would expect to find the dermal membrane also contractile. It is capable of contraction to a slight.extent under certain conditions. The chief contractile fibers, however, seem to be found only in the sphincters of the chimneys, which probably work against the general pressure of the water going out through the chimney. Although in most cases, the responses of the fresh-water sponge were similar to the responses of Stylotella, there is one noticeable difference. This sponge responds to stimuli much more slowly, and the rate of transmission, where any at all is observable, is much slower. BIBLIOGRAPHY Annandale, N. *rz1 Note on the Fresh-water Sponge, Ephydatia japonica and its Allies. Proc. U.S. Nat. Museum, Vol. 38, pp. 649-650. Balfour, F. M. 79 On the Morphology and Systematic Position of Spongida. Quart. Jour. of Micr. Sci., Vol. 19, pp. 103-109. Delage & Herouard. 200 Zoologie Concrete (2), 1 Partie. Evans, Richard. 299 «6(The Structure and Metamorphosis of the Larva of Spongilla lacustris, Quart. Jour. Micr. Sci., new series, Vol. 42, pp. 363-463. Grant, R. E. 225 Observations on the Structure and Functions of the Sponge. Edinburgh Philos. Jour., Vol. 13, pp. 94-107, 333-346; Vol. 14, pp. 113-124, 336-341. Ed. New Philos. Jour., Vol. 2, pp. 121-141. Jaffe, E. ‘72a Bemerkungen uber die Gemmulae von Spongilla lacustris L. und Ephydatia fluviatilis L: Zoologischer Anzeiger, Vol. 39, pp. 657-667. ’12b 6©Die Entwicklung von Spfongilla lacustris L. und Ephydatia fluviatilis L. aus die Gemmula. Zoologischer Anzeiger, Vol. 39, pp. 705-719. Minchin, E. A. 200 Sponges. Lankester’s Treatise on Zodlogy, p. 44. 166 GEORGE T. MCNAIR. Parker, G. H. ’I0 The Reactions of Sponges, with a consideration of the Origin of the Nervous System. Jour. of Exp. Zo6l., Vol. 8, pp. 1-41. } Potts, Edw. 787 Contributions toward a Synopsis of the American forms of Fresh-water Sponges, etc. Proc. of the Acad. of Nat. Sci. of Phil., Vol. 39, pp. 158-279. Sollas, Igerna B. J. ’00 Sponges. Cambridge Natural History. Vol. t. Wilson, H. V. 710 A Study of Some Epithelioid Membranes in Monaxonid Sponges. Jour. of Exp. Zoél., Vol. 9, pp. 536-571. pte STUDIES IN MARINE ECOLOGY: I. THE DISTRIBUTION OF COMMON LITTORAL INVERTEBRATES OF THE WOODS HOLE REGION. W. C. ALLEE, MARINE BIOLOGICAL LABORATORY AND THE UNIVERSITY OF CHICAGO. The distribution of animals within the Woods Hole region has been well studied by men interested in individual species and by those concerned with general faunistic problems. Some aspects of the ecology of the region are thoroughly set forth by Verrill and Smith in their “Report on the Invertebrate Animals of Vineyard Sound”’ made fifty years ago. This classic study re- mains the best account of the ecology of littoral species available. The extensive “Biological Survey of the Waters of Woods Hole and Vicinity’’ ! completed about ten years ago by Sumner, Osburn and Cole, while a mine of information concerning the animals of the region, was directly concerned with dredging operations and has little to say at first hand concerning the animals of the intertidal region or those found just below the tidal zone. Among other suggestions they recommend (p. 25) that the intertidal fauna should receive the same detailed atten- tion that they have given to the bottom dwelling species. In the absence of a report by the person best qualified to write on the subject, Mr. George M. Gray, the present series of papers has been prepared to make available information accumulated in nine consecutive summers’ experience with the inshore inverte- brates of the region. The work has been done in connection with a teaching appoint- ment in the course of Invertebrate Zodlogy of the Marine Biological Laboratory. It represents the collaboration of eigh- teen staff members and of about four hundred students. Many of the present collecting methods were installed in conjunction with Professor Caswell Grave, my predecessor in charge of the 1 This report will be referred to hereafter as the Biological Survey. 167 168 WwW. C. ALLEE. course, but the records here used have been kept from the begin- ning by myself with the occasional help of other staff members.! The records are based on the bi-weekly collecting trips of the Invertebrate Class and cover most thoroughly the period from about June 20 to August 15. These trips have been supple- mented by expeditions made by instructors and by special trips for particular observations. The organization of field work for eight years has been to divide the class into as many collecting teams as there were instructors. One person from each team was appointed recorder for the day and was supplied with a list of all the animals previ- ously taken from the locality under consideration. The animals found were recorded according to habitats. The complete list for the year was made up from these combined records.’ 1 The following people have been at one time or another members of the in- structing staff of the Invertebrate Course and have contributed to the data on which this series of papers is based. Without their codperation this work could not have been done. Caswell Grave, Raymond Binford, E. J. Lund, George A. Baitsell, T. S. Painter, F. M. Root, W. J. Kostir, Robert H. Bowen, C. L. Par- menter, G. S. Dodds, Robert Chambers, Jr., Ann H. Morgan, W. J. Crozier, Donnell B. Young, J. P. Visscher, J. A. Dawson, Christianna Smith and E. A. Adolph. I am indebted also to Mr. G. M. Gray for much valuable aid and friendly assistance; to Dr. Mary J. Rathbun for identification of the Brachyura; to Mr. Waldo L. Schmidt for similar service with the Anomura and Macrura; to Mr. Clarence R. Shoemaker for similar service with the amphipods and isopods; to Professor E. S. Morse for assistance with some of the molluscs and to Professor Raymond Osburn for assistance with the Bryozoa. 2The formal record of collecting experience has been recorded in abbreviated form on library cards which are deposited in the Library of the Marine Biological Laboratory. An annotated catalog of the distribution has been prepared as Study II. of this series and deposited with the library of the U. S. Fish Commission who have kindly agreed to furnish copies to the libraries of the Marine Biolog- ical Laboratory at Woods Hole; the Museum of Comparative Zoédlogy at Cam= bridge, Scripps Institution at LaJolla; the United States National Museum at Washington and to the Harpswell Laboratory at Mount Desert Island, Maine. The catalog shows the littoral invertebrates collected during the years I915—192I1 inclusive. Each locality in which an animal has been taken is recorded. The number of years which it has been found in a given locality is shown and an index figure of comparative abundance is also given. Where possible and desirable the location of particularly favorable collecting grounds is given with. some exactness. This elaborated catalog forms the basis from which the facts presented here are drawn and together with the present report gives the background for the two following studies. ee eee = oe STUDIES IN MARINE ECOLOGY. ‘ 169 So far as possible, identification was done in the field. Doubt- ful specimens were referred from one instructor to another. Specimens new to the locality or difficult to identify were brought into the laboratory for further study. With the exception of the arthropods, few of the specimens have been referred to experts although we have gradually accumulated a type collection of animals found. The identification of animals has been made on the conservative basis that when doubt existed, the specimen was referred to the more common species. Wild identifications have been eliminated as far as possible, even to the extent of throwing out the entire reports of inexperienced instructors. In spite of this care, mistaken identifications have probably been turned in and accepted. The list here given is substantially correct since the animals have either been reported by qualified collectors or placed on the list from demonstrated specimens: The imperfections lie largely in failing to distinguish closely related species and in possible errors in distribution records. II. The collections upon which this series of reports are based have been made largely in the littoral zone as defined by Edward Forbes; that is, between high water and a depth of two fathoms. This is not the littoral zone of modern zodlogists, but the term has been used with so great a variety of meanings that the extent of the study can be more easily and definitely located as being in the intertidal and ad- or sub-tidal regions. The intertidal zone is much restricted in the Woods Hole region on account of the slight rise and fall of the tides. The 1 Murray and Hjort use the term “‘littoral zone”’ to include the region near the shore down to a depth of 30 or 40 meters: ‘‘almost as far as there are sea-weeds.’’ It is frequently used as by Petersen to include the entire continental shelf. The botanists tend to be more exact. Kjellman limits the term to the region between extreme low and extreme high tide. Davis regards the littoral zone as extending from about mean low water to the highest point at which alge can grow. Flattely and Walton (’22) follow Cotton (’12) and define the littoral region as extending from the level of highest marine vegetation, to low water at neap tide. I prefer to use littoral in its original meaning of ‘‘pertaining to the shore’’; intertidal or tidal zone adequately and exactly describes the region between the tide lines, and sub-tidal or adtidal are exact terms, if not the most correct etymolo- gically, for the region below low tide. The question is discussed in the Biological Survey, Pp. 179. 12 170 W. C. ALLEE. tide tables of the U. S. Bureau of Commerce show a spring tide range of about five feet for this section of Buzzards Bay and only about two feet for Vineyard Sound stations. Studies have been made in the following localities: WHARF PILINGS. Crane’s Wharf Pilings—This is a comparatively new wharf located near the public steamboat wharf in Great Harbor at Woods Hole. At the shore end the water at low tide comes close to the retaining wall and some collecting has been done annually in the crevices of the wall. The water at the outer end is over twelve feet deep. The number of species and of animals on these pilings has increased noticeably during the period of observation. Vineyard Haven Wharf Pilings—The old New York and Portland Wharf on the south side of Vineyard Haven is located well out toward the Sound. The water here comes.up on a sandy beach which at low tide is bare for a considerable distance under the wharf. At the outer end the Government Chart shows II feet of water. In my experience the water is deeper. This is an old wharf with many pilings rotted off below water level. Some of the pilings are reproduced in the American Museum of Natural History in New York. Collections from both wharfs were made from boats by means of the usual scrape nets. Marine Biological Laboratory Pier on Glass Slides—F¥or a number of years, glass slides have been placed under the M. B. L. pier in connection with other studies. In 1921 the slides were carefully examined by Dr. D. B. Young and myself after they had been suspended in water under the pier from July 1 to August 9 at a depth of about six feet. The M. B. L. supply float containing animals from all parts of the region was only a few feet away and accordingly more species were attached than might normally be found. RocKS AND FLATs. Hadley Harbor, Southwest and Southeast Gutters —These natu- rally narrow rocky gutters have been further narrowed artificially 1 Consult map. - BULLETIN. 70)50 rae? on Gon! (Gt > Ps Aa = \ Nour | Mil JNolurice es 9 MONUMENT BLACH ' ° ' Topvs i a ORTH FACMOUTR z INVES NECK, en “ we s ‘0G 1 MARBOR rg BUTERS iL. Ve WEST FALMOUTH \ KA Ke ‘2 te) lm, > On, FALMOUTH HEIGHTS. Been PENIKESE 1 % Ae % ot sows” “ARD SOUND SSCL yard Haven Wharf. 7. Statia on I, Vineyard Sound; near oyster beds. 8. Statio on 2, Vineyard Sound; Chetopleura grounds. 9. Kettle on 3, Vineyard Sound; Amarecium pellucidum grounds. ro. North cality 1 of the dredging in Great Harbor is just within the 11. Ganse rbor entrance. t2. North e’s Wharf. Localities 2 and 3 in Harbor dredging are nearby. 13. Squete¢ ey’s Bay and Sound Gutters. The figure stands approxi- The mé itely on Blind Gutter Bar. | | . | | poiAL BULLETIN. VOL. XLIV., PLATE 1. MAP Nouticol Miles SHOWING PRINCIPAL LOCAL GEOGRAPHIGAL. FEATURES REFERRED TO IN ACCOMPANYING REPORT BASED ON US C¥ ES. CHART NO. tie Wevannis COnTeRvEce PORTH FACwouTR SY oy os -~ o g Ve o ipa 0 as ge ¥ O/SNOP RCLE RAS Jwoae OLY sUcconmesser swose eT VESSEE, incr hr CROSS MMT VESSEL = a ens WACae seure “Crt crareacuioocx fl ' é Vineyard Haven Wharf, ' Station 1, Vin ' Station 2, Vin ' Station 3, 7. Station 4, Vineyard Sound; starfish hole. eyard Sound; near oyster beds. 8. Station 5, Vineyard Sound; sand dollar bed. eyard Sound; Chetopleura grounds. 9. Kettle Cove. Vineyard Sound; Amarecium pellucidum grounds. 10. Northwest Gutter Flats. Locali : Orality of the dredging in Great Harbor is just within the 11. Gansett Bay. harbor entrance, Q 12. North Falmouth Flats. : wee batt. Localities 2 and 3 in Harbor dredging are nearby. 13. Squeteague Harbor in the North Falmouth Flats complex. Bay and Sound Gutters. The figure stands approxi- The map is taken from Bull. 31, U. S. Bureau of Fisheries. ately on Bi nd Gutter Bar, hed r pecs Danone a9) ae ant Ganesan Bae A eae eee ’ $i) pe THRE ieee v t be a es 1 my . 4 ‘ a Sis yore epee aN A | eM ‘ ‘ y ; CONG Neue >it g . A , j ty i s, Giuss sides boats i ae | ection, with opie atin 2 Sk | . : ve: : os 1 , pa . ys ined by De B 4 “4 | ‘ Be NAR OS i Cheeprt th of about: x anithals Prom) ally pea canine lion ape tnilcabeatlens = went a etal rem Tn Tee ce. miuprayy alk 4 : sie ‘ STOUR nn nae Beam eee De sad eennay thy Iocan eS a — ’ BAyy Fed a’ ae - dod dehsare PANES Prayer .p aise igunct. ‘ yatlob bese jbauoe Pasyoniv .2 adiseie -.8 670) aed oc ROCKS AND . ebsvotm! gaat tasiut) seowilsioM,.o1, . ebanot rousing see ak jhe volt sy9aiee HY si Pic a ei ose i ielh divomila BSETteT 3 haste Dea . wentatnteewt al sodinH sugestoupe .g% sai ath psn iY op Meh taord niedet af qam sat soda i é bas t ith at agg y baw STUDIES IN MARINE ECOLOGY. 171 so that they can be bridged. Strong tidal currents run through them the greater part of the day. They are relatively shallow, rock-walled channels, containing about six feet of water and are connected with open water by creeks which are also rock edged. Small patches of mud and sand occur frequently and the whole system of protected waterways supports many plants, Asco- phyllum, Fucus and Sargassum filapendula. Hadley Harbor Flats, Northwest Gutter—Northwest Gutter separates Uncatena Island from Naushon. Before opening into Buzzards Bay it enlarges to form an approximately square expanse of shallow water about 250 yards along the south and west sides and about 400 yards in greatest diagonal. In most places the water is so shallow that it is difficult to push a boat along at low water. The sand bar over the channel of the gutter is fully bare at extreme low tide often to the extent of an acre or more. The wide channel is kept scoured clean by the current, but behind the protecting sand and gravel spits, organic debris has accumulated to the depth of several feet and supports a rank plant growth composed chiefly of eel grass. At the Bay entrance there is the usual accumulation of rocks which extend off to a sand bottom some four feet below the lowest tide. The mud flats are bordered by rocks partially buried by mud. Gansett—Gansett is an offshoot of Quamquissett Harbor and has the same opening into Buzzards Bay. The main axis extends at an angle from the opening so that the back portion is usually protected from the direct drive of the waves. The opening is about 200 yards wide and the bay is approximately twice that length. At mid-mouth at mean low tide the water is 18 feet deep. The sides slope in rapidly near the shore so that there are only narrow Strips of the different habitat zones. At the sides are the customary rocks and the outer corners are guarded by rock piles. At the head of the bay the shore is sand mixed first with gravel and lower with mud. Eel grass comes within two rods of the water’s edge at low tide and thickly covers the bottom throughout its extent. North Falmouth—tThe collecting grounds here are scattered. They are located at the head of Cataumet Harbor and extend 172 W. C. ALLEB. over Squeteague Harbor which opens from the former by a winding narrow passage. Except for the dredged passage, most of the region can be waded at low tide. Much of the ground in Cataumet and almost all in Squeteague Harbor is left bare by the spring tides. The collecting is over a wide range of bottom: sand, mud, scattered rocks and gravel with and without eel grass and other sea-weeds. There is a somewhat sparse collection of rocks along the shore line. Lackey’s Bay.—Lackey’s Bay belongs to the Hadley Harbor complex. It is located on the Vineyard Sound side between Naushon and Nonamasset. The part studied forms an ex- panded entrance to Middle Gutter which, by the construction of a causeway, has become Blind Gutter. The current is much diminished by the causeway and the inner part of the bay is deeply overlaid with muck. Eel grass is abundant. The region most studied is about 400 yards long by 200 yards wide and is separated from the Sound by a sand bar which is left bare at low tide. DREDGING. The dredging has been largely in three localities in Vineyard Sound. These are the sand dollar bed (Map, No. 8) near the east side of the entrance to Tarpaulin Cove in about 20-30 feet of water. The bottom material brought up by the coarse dredge used is largely composed of shells. The starfish hole (Map, No. 7) is further east and still off Naushon, has about 90 feet of water. The Chetopleura grounds (Map, No. 3) off Nobska have about sixty feet of water. The bottom is decidedly pebbly. Some dredging has been done further east on or near the planted oyster bed (Map, No. 2) in Falmouth Harbor. The bottom here is sand and gravel in about 60 feet of water. In 1921 we dredged off the west entrance from Vineyard Sound to Great Harbor (Map, No. 4) in about 80 feet of water. This is over an Amarecium pellucidum bed. In 1920 we dredged in Great Harbor (Map, No. 5): at the east end of Nonamasset in 10-12 feet of water; in the Fish Commission Hole at a depth of 50 feet and at the West end of the passage in Woods Hole in about 20 feet of water. The dredging work has been largely incidental and the results are given chiefly as a means of comparing the more extensive CTA Ne eh a gee RETRO Ra < (V. ete STUDIES IN MARINE ECOLOGY. 173 results obtained by the dredgings of the Bzological Survey with our main work further inshore. A Hapitat CuHecKk List OF THE COMMON INVERTEBRATE ANIMALS OF THE Woops HOLE LITTORAL WITH DISTRIBUTION RECORDS FOR 1920 AND 1921. The appended list of animals is based on all the collecting ‘done since 1912. The distribution records are based on the reports from operations in 1920 and 1921. ‘The statistics given are from team records. -Thus in these two years, two teams collected Chalina from the mud, eight teams have recorded it from rocks, ten from wharf pilings and nine from dredging. The figures given show no indication of the number of specimens taken other than that suggested by the fact that the more animals present, the greater the probability that all teams would find them. Anyone interested in the abundance of these animals in particular localities is referred to the second study in the present series. The tabulation is from the reports of 30 collecting teams operating on wharf pilings; 52 from sand, mud, gravel and eel grass; 56 teams from rocks and 42 from dredging. The records of plankton have been kept in a different way and the presence of recognized animals in late July or early August is merely checked. The classification of habitats in the field has sometimes been left to the judgment of the student recorder and it is entirely probable that some of the 52 recorders thought a given habitat was best recorded as “‘sand’’ while others regarded animals from ‘ a similar place as “‘mud’’ dwelling. All gradations between the two exist and the conditions under which the collecting was done do not permit a more refined grading. The error arising from this source is somewhat compensated by the fact that no dragnet collections were made as it was desired to find where individual animals live as well as to collect different species. . Unidentified animals have not been included in the habitat list unless the genus, at least, could be determined with some assurance. All the records are for living animals since for eco- logical purposes the recording of dead shells can only be worth- less and confusing in a region where tidal currents run strongly and where the shore birds distribute shells even over the land. 174 W. C. ALLEE. The nomenclature follows Pratt wherever the species are listed in his Manual. Other species have the name given them in the catalog of the Biological Survey. No attempt has been made to give synonyms since these can usually be found in the Survey catalog. The arrangement of species within the major divisions is | alphabetical. While this does violence to all principles of taxonomy, the taxonomic sense is not strongly developed at present, and this method renders the material more easily available to the average zodlogist than would be the case if a strictly taxonomic system were followed. - The records given in the habitat list are necessarily abbreviated. Thus, Hydractinia is recorded as taken from “‘sand’’ when the whole record should read: “on shells inhabited by hermit crabs taken on sandy bottom”’; or other animals, as Sagartia lucie, recorded from “‘mud’”’ which does not mean that the anemone was growing on the mud but that it was found attached to a bit of board or rock surrounded by typical mud conditions. The classification headed “eel grass’’ includes records of animals living free among the eel grass, as Pecten; attached to eel grass, as Pennaria; crawling over it, as Ophioderma; on the substratum at its base, as Mucrocione; or burrowing in the substratum at its roots, as Cumingia. In addition the lumping is still greater for one must remember that eel grass begins to grow on fairly pure sand and extends back to the pure muck of the flats. A number of animals are recorded under “rocks and rock- weeds’’ which were taken only from the substratum under or among the rocks. Whenever all the records are for an animal so found, the entry has been appropriately labeled. Under ‘‘range”’ is listed the information at hand showing the distribution of the animal along the Atlantic Coast. The abbre- viations are: N., north ranging; S., south ranging; M., approxi- mately mid-range; L., local; C., cosmopolitan. Whenever an animal is known to extend twice as far north of Woods Hole as south, it is listed as north ranging and vice versa. (Cf. Hoyle, 1889). Some relations between the local and geographical distribution will be discussed in a later paper. be Rot one Deen. STUDIES IN MARINE: ECOLOGY. 175 A go ; 80 S| SUS | ae eee slog |g 3 leet 1e) v oe) Ay = oH ; & |e m4 i) pu Species. I a 4 g No. Teams Reporting. =) 52) }) 52) 5 2n5e2 56 30 | 42 PROTOZOA. List not given. PORIFERA. Chalina arbuscula.......... Si. 2 8 I0 | 9 pECizonancelatan(@)ern see S. 2 it 33 25 |16 Granivarciiiaiaensn. tee ee N. I II 30 |13 Leucosolenia botryoides...... N. 13 Gy || at Microciona prolifera........ Ss. I 4 43 30 RCL CNS. CULE yee eho Ij. I ir CG@LENTERATA. Hydrozoa Abietinaria abietina......... N. I2 Bougainvillia carolinensis....| S. I 2 I N sp.? Campanularia calceoltfera.... I Campanularia sp.?......... I 5 Clava leptostyla............ N. I (CUI Os re ROC ee Ie. 2 Clytia bicophora............ N. I I I N sp.? © Clytia cylindrica........... I Eucheilota sp.? medus@...... : N Eudendrium album......... 2 4 6 Endendrium ramosum....... M. 4 22] \ Eudendrium tenue.......... I Gemmaria gemmosa......... Si 5 ee I I N Gonionemus murbachit....... lee Hydractinia echinata........ N. 9 |18 | 3 5 12 II Obelia bicuspidata..... Neer ss I I Obelia commissuralis........ St I 1 8 ip Wy 2 || Ni Goats Obelia geniculata........... N. 5 6 “a Pennaria tiarella........... Ss I4 4 Io | 4 N Phialidium sp.? meduse..... N Podocoryne carnea.......... N. 2 I Schizotricha tenella.......... Ss I wir | 2 Sertularia pumila.......... N. | II 30 I2 |10 Stylactis hoopert............ Se Tubularia crocea........... Ss. 30 | 7 N Scyphozoa. Aurelia flavidula........... M. Cyanea capillata arctica..... N. Daciylometra quinquecirrha...| S. Haliclystus auricula......... N. 176 W. C. ALLEE. : = : me Pee B Sy a BS | Dae edhe ae | ie aaa & (Se Pe Gia) |i & E 5 a wn Oo oO 9°90 Ay =I es} ; m |e cs Ao Q cy Species. S % fa g ; No. Teams Reporting. ‘ | 52 | 52] 52] 52 56 30 | 42 Anthozoa. Astrangia dan@............- S: eel II 4 Alcyonium carneum......... N. Etdwardsia elegans.......... N. it | & iT Flloactis producta........... S: 2 Metridium dianthus......... N. 29 24 Sagartia leucolena.......... Sh Bail.3 B 37 Ir |.4 ‘ Sargartia luci@..........-.. S- BN) 6 38 6 | 2 OLOTIUG MMOGESEG vale ies el S: 4 I2 under PLATYHELMINTHES. Turbellaria. Bdelloura candida.......... Shes S i 2 7 Bdelloura propinqua........ I | Polycherus caudatus........ N. i as} 5 2 ro Procerodes wheatlandi....... N. I Stylochus ellipticus.......... N. B22 | x I 6 2 SVMOMUIS Bits acoaccacaoee IL, I I Syncelidium pellucidum..... 2 Nemertint. Amphiporus ochraceus....... Ss. 5 Cerebratulus lacteus......... Ss Ge Navan 2 2 N under Cephalothrix linearis........ N I under ETMCUSUS Debus, hee cians ef eee age 6 5 I 2 WEANeUS OLCOlOP. wae che coe nc Ss mo a it 2 Miacrura leidyt... 3. -...+..-- M 6 |I5 8 3 under Tetrastemma vermiculum.....- S) Bales NEMATHELMINTHES. Pontonema marinum........ | Ss. | | i |) a | | 5 | 6 | I9 | 6 > ECHINODERMA. Asteroidea. INGUG ATS HORUS ob oon oD Oe Sc Sh r| 8 4 | 28 I4 |29 N Asterias vulgaris,.........+. N. 3 I/4 Henricia sanguinolenta......| N. : i 6 4 |II STUDIES IN MARINE ECOLOGY. Wa) E ay. ; ; : GB a9 S 0 (| Sees ie. | ae ey |) s gislalaio| #2 12/21 3 . "D uw a g|% Pla | ae |] ale Species. I 2 fo g No. Teams Reporting. fas | ens 52 | 52 | 52 | 52 56 30 | 42 Ophiuroidea. Amphipholus squamaia...... N. B} Pe Ophioderma brevispina....... Shs 5 Echinoidea. Arbacia punctulata.......... S: 6 | 18 a Wins) N Echinarachnius parma....... N. I 12 Sivongylocentrotus drabachiensis............. N. Holothuroidea. Leptosynapta inherens...... Ss II |24 ro Thyone briareus............ Se 25 6 2 ANNELIDA. Archiannelida. Dinoprilus Spite 2... 22s) | N Polycheia Ampharete setosad........... Se Amphitrite attenuata........ I 9 Amphitrite brunnea......... N. I Amphitrite ornata.......... S: Io | 8 3 2 2 under Arabella opalina...........- S. @ (x7 | 2 5 I iw | under Arenicola cristata........... Si 3 6 I under Autolyius cornutus.......... N. I 9 Autolytus varians........... Ss. I Chetopterus pergamentaceus..| S. I Cirratulus grandis.......... Ss. in | @ | x 5 I under Cirratulus tenuis........... I Capitellaspenneeee ie ee Clymenella torquata......... M. I7 |25 6 6 : under Diopatra cuprea............ S; 6 |14 2 I 2 under Drilonereis longa........... it |) oy i Enoplobranchus sanguineus ..| M. 3 Glycera americana or Ss: oe aol I dibranchiata............. M. 33 under Harmothe imbricata........ N. ey 5) |) oat 3 26 27 |Il 178 W. C. ALLEE. Dh ob en Ieeretete | Cees I) 13) d Z/2/418] z | oo a) eeumes Aig % | 3 a | ae as Species. 5 3 4 2 No. Teams Reporting. =) 52 |52)|52| 52 56 30 | 42 Polycheta Con’t Hydroides hexagonus.......- Shale Sei] 2B |x 3 39 L722 IGQonice viridis... .......+-+- Se Be Spe Lepidametria commensalis....| S. I : among Lepidonotus squamatus...... Nea as | B.S 2 2 37 29 |22 ELEprea Tuva. .....-+--+--- Ss. I 4 under Lumbrinereis hebes........- Lumbrinereis tenuis......... Si 4 \rQ-| 2 I 3 I4 under Maldane urceolata.......... St 319 2 Marphysa leidyi............ I 3 IGRAE WG IOHIOS 3506 o00000 46 Ss WA | GO) 2 3 8 3 Nerets pelagica............. N. 4| 6 9 29 |22 Nisp.? INIEKEUSTUUN ENS «stele eel siewene oie is N. I7 |20 9 2 I under Nicolea simplex...........- Toes 2 7 102) IG Pectinaria gouldi........... S) Cie eT 5 Potamilla sp.?............-. I PEROLOERS Dettecyers isin) ousstaekorene eines Phyllodoce catenula......... INS || |B 3 5 I7 |I3 LENG (HUT OCHOS Goo nn Oba soot S) led I Platynereis megalops........ S) I I Podarka obscura............ Ss 2 4 4 7 || © Polycirrus eximeus.......... Ss: IO |14 4 I5 6 |12 IZUMI Soa d oo vo Deo Be b Ee 2 Sabella microphthalmia...... Se was 3 WH | a Sabellaria vulgaris......../. Sal eee & 6 Boi fi SCOLOPIOSOCULUS EE nent ae i623 || a 6 Scoloplos fragilis........... Ss. 8 |I9 9 I : under Scoloplos robustus........... Ss. 8 |I0 3 I : under SPLOM(SELOSAM)s os meeeeces sec I Spirorbis spirorbis.......... N. 3 |) 4k || Bo) ara 26 IQ Spirorbis tubeformis........ = I I Sthenelais leidyi............ S: 6/5 4 3 ‘4 | 6 Terebellides stremt.......... N. Thelepus cincinnatus........ N. I 3 Trophonia affinis........... S 4/1 6 Chetognatha and Sipunculoidea Phascolosoma gouldii........ M. I3 |22 3 2 under SOSULGISD ates tees tee eee N wT actA TR aig STUDIES IN MARINE ECOLOGY. 179 w Es 3 3b | 5 fa in So oO q to) | JJEIZIE16] ge |e |S) & Fee S| at ee et Soc eemsia lt 5 3 | s OS | eee) 4 Taal tary Species. a 2 % ie No. Teams Reporting. =) - 52 56 30 | 42 Rae ear ak tenn Os 5 10 Ti MOPS PACT eres 3) cc ER ORerS scene 2 | 2 Aachen 3 I Sidsener Olecherone I Siena Minko : Ty al geeel Spence Men eneterehs are 5 neal aay 31 |14 NX se re ee : I nic Os ee ee : 4 21 18 |12 owen aes I 22 So eOee I Hien ACER EE I | ee eaaete : : 10 | 14 6 | 6 MR e ascetics I 2 : I ‘ “Membranipora pilosa........ 6 21 13 | 5 . Membranipora tenuis....... 6 S Microporella ciljata......... € ES Bie 2S = Schizoporella biaperta....... 5 a Schizoporella unicornis...... So ae Ga 8 32 25 |Ir ih Smittia trispinosa nitida..... M. I 3 I5 i ARTHROPODA. Phyllopoda. Evadne nordmanni.......... M. N Podon leuckartt............ M. ae N Cirripedia. Balanus balanoides......... M. i ees 2 48 2 NX Balanus eburneus........... Ss. 1 |B 6 18 IQ |I7 UGHOS CLOG 6 ooo bee Kae (Oran a I Arthrostraca. ‘i Amphithe rubricata......... N. Tha] ae 6 6|8 Autone (Lembos) smithi..... L. I 5 5 | 9 Aiginella longicornis........ N. I 4 , *Caprella geometrica........ Si. Tones 6}. + 23)+s| 26.) 5 NX ‘ Chiridotea c@ca............ S 2|2 2.|-- 2 Corophium cylindricum...... 2 Cyathura carinata.......... N. I I Edotea triloba.............. IMI, |) @ |B 2 2 * In part Aginella ‘p vd 180 W. C. ALLEE. Unclassified. Eel Grass. g Species. I % Arthrostraca Con’t. Erichsonella filiformis....... N. Gammarus (Sev. sp.).......- 2 Haustorius arenarius........ S: Idothea baltica............. M. Idothea metallica........... Sis Idothea phosphorea......... N. TLD MGTLWG «scale sere cle ee IN || 2 Eigydaioceanica............ N. Onchestigvagtlisy. s. .- 5-6. sss S. |ro Spheroma quadridentatum ...| S. Talorchestia longicornis...... Ss. WOH ADSICQUOLLNID em eee 2 ene N. Unicola irrorata............ N. Thoracostraca. Callianassa stimpsoni....... Ss. I Callinectes sapidus..:....... Ss. ip | oe Cancer, borealis. . 22. s-+ sss N. 3 Cancer trroratus...........- M. Geer 4 Crangon vulgaris (Crago septemspinosus)...... M. BA 16 Carcinides m@nds.......... S: 16 | 4 4 Heterocrypta granulata...... Ss: Heteromysis formosa........ IN itn ales I Hippa (Emerita) talpoida....| S. IEIQOS: GOUPOHHHTIS ano bo bob bo N. LENO CIS Beco aut Oe Sh 14 | 8 Io Libinia emarginata......... Ss; Dae 7 Lysiosquilla armata......... Mysis stenolepis............ M. Michtheimysis mixta........ Ovalipes ocellatus........... Se Tn 33) 2 Pagurus acadianus.......... M. it Pagurus longicarpus........ Se iS} [ie@) |x 0) 2 WOUKUSEPOMIGHY IS ete eae S. | 1 | 4 |ro i Palemonetes vulgaris........ Ss: Onl 7 16 Panopeus sayi or Neopanope texana sayi.... \ Be 2) Ey Ee te ZELVEMNULIGO See eae S. I Pinnixa chetopterana....... 2 2 Pinnixa sayana (cylindrica). . Pinnotheres maculatus....... Se 3 ZOVLUMUSESOY Litericosieie eee Ss. Squilla empusa............. Ss | Rocks and Rockweed. it among 6 2 near 3 8 iT 33 20 I Io 2 q 2 I I 8 7 27 it among 6 HO 16 STUDIES IN MARINE ECOLOGY. 181 B Teles : : : Syele ie | Ge] See gisla ene tes be ie | = o |e ©) v Re) a = : to | CB ca 04 A pA Species. =I 2 mM |e No. Teams Reporting. =) 52|52)52)| 52 56 30 42 Thoracostraca Con't. WCOSPUSII ALON Wail. 1c ees: Ss: Pro |) S 2 8 (QHED: IITTUGRS oiselo Cree SI Oig Ore Se 2 Virbius (Hippolyte) zostericola| S. | r | 4 | 2 25 I Arachnoidea. Anoplodactylus lentus....... INIe | 2 4 Io | 2 PaWeneremPpuUsas.. 2522002 Shs 2 it 5 N Limulus polyphemus........ Sb |) i |e |x) | 2 6 I Tanystylum orbiculare....... Ss Io | I MOLLUSCA. Amphineurd. Chetopleura apiculata....... 1 Ss | 3 | | 19 16 Gasteropoda. Acme@a testudinalis......... IN; | B | wz I 26 r || B Bittium aliernatum.......... Mi] NO 16] 2 | eo 17 Ir | 8 Busycon canaliculatum...... Ss. | i D ISISNCOD, GQUPATE Sn Soa 5 0000p eo Ss. w | i I I Caecum pulchellum.......... 2 Cerithiopsis emersonit....... Ss |) 2 2 3 6 Cerithiopsis greenti......... Ss A || i Soke 2 Cerithiopsts terebralis........ 2 Columbella avara........... So he ees i) a 9 13 I4 |19 Columbella lunata.......... Se sesh eg I2 20 30 |21 Coryphella gymnota......... N. | 1 I w 2 I5 Crepidula convexa.......... Sy |) Au lees ac) |) ae 6 20 AL |e Crepidula fornicata......... Ss | 2 Olesya 2 21 I2 | 9 Greprdula hland..... 2.25. -. Seo Qe 7 HG. | 2 iis I4 /I5 IDOPUS WY ECIB 8 a Bee 6 oes OD N. Ww |) 3 EVISU GRIOKOLUGE ems ei-te e o M. 4 I Eupleura caudata........... Ss. BW oehe || se 4 5 WGC CUIGSULILCLO ae eee S: Ab ak || 2) awa I3 Cie || (a) Littorina irrorata........... Ss: 2 WAULGYANONLILOLEO eee N. | 1/15 | 8 | 6] 38 29 27 Littorina palliata........... N. ALN I || | aes 24 G7] | a WTO ION LOLS ieee N. yall AL |) BN) 20a) 32 6|1 Melampus lineatus.......... Ss Al |) “lh |) 2 INISSQOUSOLGIO . 5 2 ee ess ss S. 35 |14 2 9 INIGSSCMT-LULILAL Gama ete ee). S: 13 |r6 | I 3 8 ine || 3 INIDSS GRUUDE Nira on yatefete cee oils viens Sy. I Natica duplicata............ S. AW (6) |) x 3 IT among INIGLICONN CLOSE erie eine: M. 2A |e Ties er 2 182 W. C. ALLEE. = i 8 Ey £ oy) | ailei@i s ao | ds 2 Sen linet Gs oR) AY a ge Agee ew at ca Ree” if % & : & Re arsyota STUDIES IN MARINE ECOLOGY. 187 in some abunbance and by the larger number of peculiar species. They are not closely related to other habitats if one excepts the conditions under the rocks which scarcely form a different association. Its main difference is that burrowing and exposure are more limited than on the open flats. The animals common on the flats are particularly absent from wharf pilings showing, as would be expected, that these habitats have little in common. If quantitative data were available, the distinctions here found would doubtless stand forth more plain!y. The fact that a beginning can be made in indicating relationships by the methods used when no relationships appeared from an analysis of bare check lists emphasizes the need of quantitative studies in animal ecology. This need was first recognized by Forbes (1907) who devised a mathematical formula for determining the existence of an association. In 1911, as the result of studying seasonal succes- sion in ponds, I concluded that qualitative work gives insufficient basis for exact conclusions. Shelford in 1915 repeats the formula of Forbes, and Michael (16, 21) has done more than anyone else in America in showing the fundamental need of quantitative investigations in ecology and in developing formule to enable: one to study associations on a quantitative basis. The work here presented is of course only quasi-quantitative in character but the greater clearness obtained indicates that much of the muddle of animal ecology may be cleared by the further development and the application of quantitative methods in field researches. The problem of the ecologist studying littoral distribution is not so hard as that of the plankton student where as Michael says (1921): ‘‘Granting the equivalent of the oak tree or pine tree association, the marine ecologist finds difficulty not only in describing it but even in finding it. Since he cannot directly witness such an association, he is compelled to rely on indirect evidence furnished by tow-net or similar apparatus. In other words his only recourse is to measured magnitudes and application of mathematical logic thereto.’’ For the exact deter- mination of such relations the methods here used are almost as gross as are the ordinary qualitative observations in trying to solve the relationships existing between littoral associations. 188 W. C. ALLEE. It is true that analysis of the results of preliminary collecting showed the same relations as the quasi-quantitative analysis of more complete records in the Woods Hole region. Unfortunately one cannot be sure that the animals found in such preliminary work are really the typical animals since they may obviously contain many incidental forms. In other words in a random sample one is more apt to collect animals typical of the habitat than incidental forms but he can never be sure of this without further work. V. THE EFFECT OF CONTINUED COLLECTING ON DISTRIBUTION RECORDS. In 1917 when the collecting records were first studied seriously there were 181 species in the catalog. In 1920 when a similar study was made, the catalog listed 242 species. In the interim the Sound Gutters, Lackey’s Bay and Great Harbor had been added to the localities visited. In 1917, I1 species were recorded only from the wharf pilings. The later lists show 7 species so limited but this includes only one (Tetrastemma) of the previous list. In 1917 two species were recorded only from rocks or rockweeds, while on the later list there were 10 such species including only Clava leptostyla from the preceding list. In the first comparison there were 10 animals recorded only from dredging; in the later one, 13, which includes four of those on the preceding list: Arca ponderosa, Strongylocentrotus droebachiensis, Heterocrypta granulata, and Amaroecium stellatum. On the 1017 list, 71 species were recorded from the flats only. After four more years’ work this had shrunk to 41 providing animals found in the sand under and among rocks are excluded. Of these only 17 appear on both lists. They are: Edwardsia elegans, LEloactis producta, Bdelloura candida, Syncoelidium pellucidium, Ophioderma brevispina, Chaetopterus pergamentaceous, Platynereis megalops, Scoloplos acutus, S. robustus, Spio (setosa?), Callianassa stimpsoni, Mysis stenolepis, Squilla empusa, Melampus lineatus, Clidiophora trilineata, Pecten irradians, and Tellina tenera. i ¢ nd ¥ 4 é f < ; > ; Uy ie = OS it STUDIES IN MARINE ECOLOGY. 189 On the earlier lists, fifteen species were recorded from some place in each of the four main types of habitats: wharf pilings, rocks, flats, and dredging. In 1920 this list had increased to 54. These results mean, as has already been suggested, that as collecting has proceeded, animals have been picked up in habitats in which they are not abundant. The scarcity of many of these is shown by the number of single specimen records on the lists. There is little doubt but that if the present type of collecting were continued long enough, there would finally be stray records of many of the animals found in the region from each type of habitat. Even dredging, which we have usually carried on in deep water in Vineyard Sound, yields a different type of animals and becomes more closely related to other habitats as a result of dredging records taken in Great Harbor. In some one or more of their dredging operations, the Bzological Survey found most of the animals we have taken from inshore digging. This result might be expected from the fact that some of their dredgings are recorded from less than 10 feet of water. If made at high tide, these would be almost as close inshore as our deepest col- lecting on wading and digging expeditions when we often collect out to four feet of water at low spring tides. In other words, in such a small region as we are now con- sidering, provided with strong tidal currents which aid in distri- bution, the animals tend to become widely distributed and occasional specimens will be found that can tolerate for a time conditions that are essentially unfavorable. Under these condi- tions the mere record of the presence of a species in a given habitat means very little unless there is due consideration of its abundance and duration in that locality. One is thus driven again to the conclusion of the last section, that quantitative work is necessary before final judgment can be passed in the matter of the constitution of animal associations. We have found no evidence that the long continued collecting over the same grounds by the Invertebrate Class, nor the com- mercial collecting of the Supply Department of the M. B. L. has affected the number of animals present within the past nine years sufficiently for the effect to be noticeable by the collecting methods we have used. With growing experience in collecting 190 W. C. ALLEE. each year we have broken previous records with monotonous ‘regularity, for numbers of species from most of the localities we visit. This could not have continued so long had the animals been becoming less abundant. The number of animals present in a given locality must depend more on the availability of suitable breeding places and abundance of food than upon such disturbing influences as summer collecting, particularly when the collecting does not reach all the breeding habitats of a region and there is adequate _means of distributing young stages. This conclusion is empha- sized by the rapid recovery in numbers of Arbacia after their almost complete disappearance following the winter of 1917-18 (Allee, 19) and in the face of their destruction by the thousands in the research work carried on in the Woods Hole laboratories. VI. SUMMARY. 1. Analysis of distribution records in the four major types of habitats of the Woods Hole littoral, viz., wharf pilings, rocks and rockweeds, flats, and the sea bottom in deeper water show that mere records of species present in the different habitats fail to indicate any relationship between the different types of associations. ; 2. By eliminating species known to be approximately equally distributed throughout and records for one year, only, and classifying the remaining species in terms of places where they are most abundant and next most abundant one finds: (a) The association of the wharf pilings is closely related to that of the rocks. (b) Species taken in dredging on clean hard bottom are found in next abundance on the rocks. (c) The associations of the flats are highly independent of the others in the region but continue in the mud and sand under and around rocks. (d) That some degree of quantitative work: is necessary in order to determine the relationships of animal associations. 3. Preliminary collecting in a region tends to give the obvious forms and gives similar results in analysis to the type of quasi- qualitative work described in this report. ee ae a wn as STUDIES IN MARINE ECOLOGY. I9t 4. The number of animals present in the Woods Hole region has not been noticeably affected by the intensive collecting carried on there during the nine years covered by these studies. LITERATURE CITED. Allee. "rr Ill. State Acad. Sci., 4, pp. 126-131. ’19 ©6BIOL. BULL., 36, pp. 96-104. 22 Ibid., 41, pp. 99-131. ’23 Studies in Marine Ecology: II. An annotated catalog showing the distribu- tion of common invertebrates of the Woods Hole littoral. MS. deposited in the following libraries: U.S. Fish Commission at Washington; Marine Biological Laboratory; Museum of Comparative Zoology; U..S. National , Museum; Scripps Institution at LaJolla; Harpswell Laboratory. Cotton, A. D. "12 Proc. Roy. Irish Acad., Sect. I, Pt. 15. Davis. ’zr Bull. U. S. Bur. Fisheries, 31, Pt. I, pp. 443-544; Pt. II, pp. 795-833. Flattely and Walton. ’22 The Biology of the Sea-Shore. Macmillan, 336 pp. Forbes. ’51 Edinburgh New Philos. Journ., pp. 192-263. ’59 The Natural History of European Seas. 306 pp. Forbes, S. A. ’07 = Bull. Ill. Lab. Nat. Hist. 7, pp. 273-303. Hoyle. 789 J. Linn. Soc. 20, pp. 442-472. Klugh. 718 Journ. of Ecol., 6, p. 230. Kjellman. 77 Nova Acta Regiae Societatis Scientiarum, Ser. III. Jubelbana, Upsala. (Read in abstract.) Michael. 716 Univ. of Calif. Publ. in Zoology, 15, pp. i-xxiii. 220 Journ. of Ecol., 8, pp. 54-60. Murray and Hort. *12 Depths of the Ocean. Macmillan, 821 pp. Petersen. ’93 Det Videnskabelige Udbytte af Kononbaaden “‘Hauchs”’ togter. Copene hagen, 464 pp. Summary in English. Pratt. 276 A manual of the common invertebrate animals. McClurg, 737 pp. Shelford. 715 Journ. of Ecol., 3, pp. 1-23. Sumner, Osburn and Cole. zz Bull. U.S. Fisheries, 31, Pt. I, pp. 1-443; Pt. II, pp. 547-795. Verrill and Smith. 71-72 Rep. U.S. Fish Comm., pp. 295-778. FOOD AND PARTHENOGENETIC REPRODUCTION AS RELATED TO THE CONSTITUTIONAL VIGOR OF HYDATINA SENTAs W. V. LAMBERT, W. S. RICE, AND H. C. A. WALKER. There has been a general belief for many years that cross- fertilization is beneficial to a race. Darwin studied the effects of inbreeding and crossbreeding in many plants and came to the general conclusion that while inbreeding has a deleterious effect, crossbreeding has a beneficial effect upon the races involved. In experiments with infusoria Calkins has found that if conjugation is prevented the race gradually weakened and finally died. He observed, however, that by the artificial stimuli of various chemicals in the culture water he was able to prolong the life of one race of Paramecium caudautam to the 742d generation. Investigations in regard to the effects of parthenogenetic reproduction have been carried on by Shull and Whitney with the rotifer, Hydatina senta. Both of these investigators have found that continued parthenogenesis in pedigree races of rotifers has resulted in a gradual weakening and loss of vigor, and in some experiments death. In Whitney’s experiments one race of rotifers died out from general exhaustion in the 384th generation and another in the 546th generation. A third race, discontinued in the 443d parthenogenetic generation had also shown a marked decrease in constitutional vigor. Investigations of other workers, however, do not seem to confirm these general results. Woodruff, in his observations on the Paramecium aurelia, found that a race of these animals kept in cultures made from natural pond waters did not undergo marked depression or physiological changes. This race at the end of eight years, after having passed through 5250 generations without conjugation, was in a normal condition. Another race, however, subjected to the relatively constant hay infusion cultures generally employed, died after passing through several hundred generations. 1 Studies from the Zodlogical Laboratory, The University of Nebraska, No. 134. 192 FOOD AND PARTHENOGENETIC REPRODUCTION. 193 King, also, in experiments with albino rats has found that continued inbreeding of the rat, when subjected to a varied and well balanced diet, does not weaken the race. On the contrary, the strains were larger, more fertile, and lived longer than many strains of stock rats in which no inbreeding was allowed. Castle, also, found after inbreeding the fly Drosophila (brother and sister matings) for fifty-nine generations, that no apparent decrease in the size and vigor, as compared to those with which he started, was noticeable. Noyes has recently finished the 250th partheno- genetic generation of the rotifer Proales decipiens and has detected no noticeable decline in vigor. Moreover, males of this species have never been seen. East and Jones are inclined to believe that if proper changes in the diet had been made in the experi- ments of Whitney and Shull upon the rotifer Hydatina senta that the normal vigor of the races would have been maintained. PROBLEM. The following experiments with the rotifer, Hydatina senta, were begun in order to determine if possible whether the gradual loss of vigor is due to the parthenogenetic reproduction or to a too restricted diet. The various races used in these experiments were kept under identical conditions as to temperature, light and culture water. Race A was fed.a colorless Polytoma diet, whereas race B was fed a diet consisting of the Polytoma plus several kinds of green protozoa, predominately Chlamydomonas. The plan has been to keep the races in a parallel series and to study the effect of the two diets upon the two races. A third race (C) which has been obtained from wild rotifer cultures in the natural environment has been introduced into the series in order that the rate of reproduction of the rotifers under restricted and controlled environmental conditions might be compared with the rate of reproduction of those under the uncontrolled conditions of nature. Acknowledgment is due Professor David D. Whitney for his suggestions and supervision of the work. METHODs. Some dried fertilized eggs of the rotifer, Hydatina senta, were collected at Lincoln, Nebraska, on September 17, 1920. One of 194 W. V. LAMBERT, W. S. RICE, AND H. C. A. WALKER. these eggs hatched on September 29, and on October I two parthenogenetic sisters were isolated. One became the mother of what has been called race A and the other became the mother of what has been called race B. The two sister parthenogenetic races A and B were kept in syracuse watch glasses. Usually once in forty-eight hours five daughter-females of each race were isolated, each daughter-female being placed in a separate watch glass. They produced the young females of the succeeding generation. Race A was fed on a pure culture of the flagellate, Polytoma. This food was grown in horse manure solutions, which was prepared as follows: eight hundred cubic centimeters of fresh horse manure and one thousand cubic centimeters of tap water steam sterilized for one hour. This was cooled and diluted in the proportion of one part horse manure solution to two parts of boiled rain water. Polytoma were then put into it in an open pan and placed at about 18° C. The Polytoma grew very quickly and in 24-48 hours immense numbers of them were produced. This culture, ranging from 24-96 hours old, was decanted and replenished daily—one part culture medium to two parts boiled rain water being added. From the surface of this culture the Polytoma were taken with a sterilized pipette. This food was centrifuged so as to make it concentrated enough so that one drop was sufficient for each watch glass containing a female rotifer and a small amount of filtered rain water. Race B was fed both from the pure culture of Polytoma and from two or three cultures of green flage'lates, Chlamydomonas, Chloroganium. euchlorum, and Euglena. As some of the green food grew in salty cultures it was always washed with rain water before being placed in the watch glasses as food for the rotifers. During the whole period of fifteen months the two races have been conducted in parallel generations. The external factors and environment have been as near alike as it has been possible to make them. The individuals of each generation were isolated at the same time, were put into the same kind of watch glasses and with approximately the same amount of filtered tap water, the only difference being in the food. The watch glasses were stacked side by side. During the day the room was lighted by the north light. FOOD AND PARTHENOGENETIC REPRODUCTION. 195 The method used for deciding whether the races had maintained their original vitality or whether they had decreased in their constitutional vigor has been the rate of parthenogenetic repro- duction. In order to determine the comparative vigor of the races A and B, their rates of parthenogenetic reproduction were obtained by counting the young produced by a mother during a certain period of time for.several successive generations at intervals of about thirty days. The first comparison was made by counting the young daugh- ter-females of both the races A and B from generations 22-29. The second comparison was made by counting the young in the same manner as above from the generations 43-47. The third was made by counting the young from generations 61-67. No green food was given to race B during the generations when these counts were made so that conditions would be identical at the time the rate of parthenogenetic reproduction was being determined. During the above named generations fifteen daughter-females of approximately the same size of each race were isolated. This was done in order to give a greater number from which to get the average of young females produced, thereby decreasing the chance of error. When the third and also the last comparisons were made fifteen females from various wild races of Hydatina senta were isolated and placed in separate watch glasses. These wild females have been called race C. This race was fed only pure culture of Polytoma. It was kept for only a few generations, being conducted under identical conditions and parallel to generations 65-67 and 169-175 of races A and B. This intro- ' duction was made in order to make a comparison of the rate of parthenogenetic reproduction between races A and B and various -wild races taken from their normal environment. RESULTS. The first series of observations on the rates of parthenogenetic reproduction in these two races was started at the beginning of the twenty-second generation. The results at this time as can ‘be seen in Table I., are extremely variable. Only five daughter- 196 W. V. LAMBERT, W. S. RICE, AND H. C. A. WALKER. TABLE I. No. of No. of Young Period of Race A. Race B. Partheno- Date of Females Iso- Growth Av. No. of Av. No. of genetic Isolation. lated and and Daughters Daughters Generation. Reproducing. | Reproduction. Produced. Produced. DOU Ste eaten I1/16/20 5 48 hrs. 9.22 10.25 Doreen wey 11/18/20 ® 48 7.20 ety Lic YAR ya 11/20/20 5 48 4 I Bie riicie eet I1/22/20 5 48 | 7.8 I.4 An eo teats t1/24/20 5 48 I.5 I.5 2 ee ely / 20120) 5 72 4.2 4.5 aay Se 11/29/20 5 72 13.8 15.66 20 ha Weick 12/2 /20 5 48 7 | Tek Average . .| 5 54+ 6.84 | 5.81 females were isolated every other day for making the counts on the two races. The average number of young daughter-females produced in each forty-eight hour period at this time was ex- tremely variable. The mothers of the A race during this period produced an average of 6.84 daughter-females while those of the B race produced an average of only 5.81. The two parthenogenetic races were then carried on parallel with each other until the beginning of the forty-third generation - at which time another series of observations was started on the number of young daughter-females each mother would produce in a given length of time, approximately 48 hours. As new generations were isolated great care was taken to pick uniform- sized daughters. One investigator did all of the isolating during this series of counts in order to insure the greatest uniformity. Considerable difficulty was experienced during this series of obser- vations because of the large per cent. of male-producing females. These had to be discarded for only the mothers producing daughter-females were considered in calculating the averages. The A race during this second series of observations produced an average of 7.58 daughter-females while the B race produced 8.88. Considerable variation occurred throughout the different generations, but the variations were less in this instance than they were in the preceding series of observations. A complete record of the results of this observation can be seen in Table II. The two races were then carried on for fourteen generations before another series of observations was begun. At the begin- orate tint eae FOOD AND PARTHENOGENETIC REPRODUCTION. 197 TABLE II. No. of Young No. of Females Iso- Period of Race A. Race B. Partheno- Date of lated and Growth Av. No. of Av. No. of genetic Isolation. | Reproducing. and Daughters Daughters Generation. Reproduction. Produced. Produced. A. B TGs 5. ae t/ 6/21 5 5 48 hrs 7.4 ro AAI roc a se t/ 8/2 so) II 48 6.6 3.8 JIGgEy see Lae ae r/t1/2t T4 Tie 48 6 I0.8 (es Oe eaten t/t1/2t 5 5 48 6.5 6.9 ASL ead eee 1/13/21 8 II 48 I2 Seo) MGS tal SOR RO 1/15/21 iy IA 52 7 6.6 Average.....| 48.7 7.58 8.88 ning of the 6Ist generation fifteen individuals from each race were isolated as before and placed in separate watch glasses,— one person doing all the isolating as in the preceding series. No male-producing females appeared in this series of observations and the number of young produced in a given length of time was more nearly uniform throughout than was the case in the two preceding observations. The average number of young produced by the A race in the last three generations of this series of counts was 2.66 in an average of 46.3 hours. In the B race for the three generations an average of 3.68 daughter-females were produced in the same period of time. TABLE ITI. Av. No. of Daughter- No. of No. of Young Females | Period of Females Produced. Partheno- Isolated and Growth genetic Date of Reproducing. and Genera- Isolation. Repro- tion. duction. | Race A. | Race B. | Race C. A I5%p G (C}Iit ene oes 2/13/21 15 13 48 hrs 4.84 5.84 OB are siti 2/15/21 T5 I5 4I 4.64 5-44 OBei es sae 2/17/21 I5 I5 47 2.90 Bai OA Ses BB 2/19/21 15 I5 48 4.25 4.37 OShwme ses 2/21/21 113 I5 05 44 I.15 2 Dower GOR ee 2/23/21 I5 I4 I5 48 3.56 4.21 4 Giz saegseanes 2/25/21 5 I4 I4 48 3.80 5.21 4.25 Average. . 46.3 2.66 3.68 3.53 At the beginning of the sixty-fifth generation, counts on the 198 W. V. LAMBERT, W. S. RICE, AND H. C. A. WALKER. wild race of rotifers, race C, were started. They were reared under identical conditions with the A and B races and the counts were carried over a period of three generations. The average number of young produced by these mothers in 46.3 hours was 3.53. A complete record of the results of this obser- vation can be seen in Table ITI. The two parthenogenetic races were then carried on parallel with one another until the beginning of the 79th generation at which time a fourth series of observations was started. During this fourth series of observations the A race produced an average of 7.46 daughter-females while the B race produced 8.29. Very little variation occurred throughout the different generations during this series of observations. A complete record of the results of this observation is given in Table IV. TABLE ee Av. No. of Daughter- Nove young Females Produced. No. of Par- SHES Period of thenogenetic | ,Date of Isolated and Growth and Generation. | Isolation. | Reproducing. | Reproduction. = Race A. Race B. A. B. FOV eve eeies ot 3/22/21 I5 15 48 hrs. 10.06 11.46 SOM Sein Maas 3/24/21 I5 15 48 7.06 7.53 ts San cena 3/26/21 13 I4 54 3.92 4.28 SOR Rika ays 3/28/21 15 I5 46 6.53 WBS) Beste ergs Sales 3/30/21 I5 ins, 48 : 9.73 10.93 Average..... 48.8 7.46 8.29 The two races were then carried on for thirteen generations before another series of observations was made. The average number of hours between counts in this series of observations was 65.33 hours, the increase in length of time being due to the slowness of reproduction which was probably caused by the low temperature of the laboratory. The average number of young daughter-females produced during this series by race A was 3.32, while the number by race B was 5.15. A complete record of the results of this series is given in Table V. The two races were again carried on parallel with one another until the beginning of the 106th generation. At this time they showed a very great degree of uniformity between the different FOOD AND PARTHENOGENETIC REPRODUCTION. 199 TABLE V. Av. No. of Daughter- Nooo pune pone Females Produced. 2 emales eriod o Pen re Date of Isolated and Growth and Generation: Isolation. | Reproducing. | Reproduction. ees Race A. Race B A. B. CO ob masa 4/28/21 14 I5 77 hrs. 4.13 6.53 Oi retensvcters cavers SH ajar 13 I5 70 BED 4.33 OlSia Grete nee eel ge 5/ 4/21 I5 T5 50 3.60 4.60 Average..... 65.33 ease 5.15 generations, but there was a slight decrease in the difference between the number of young daughter-females produced by the two races. The A race during this series of observations produced an average of 9.19 daughter-females while the B race produced 10.06. A complete record of the results of this obser- vation is given in Table VI. ABER Wale Av. No. of Daughter- No. of wouns Females Produced. No. of Par- Females Period of thenogenetic | ,Date of Isolated and | Growth and Generation. | Isolation. | Reproducing. | Reproduction. Race A. Race B. A. B. TOOK En acts 5/23/21 r4 13 48 hrs 9.28 10.07 IO /e Sidenote 5/25/21 T4 I4 48 8.14 10.64 MOS Men ee 5/27/21 I5 T5 48 6.60 7.20 TOO Manic erst: 5/29/21 I5 I5 48 Ir.46 I1.46 Op ose ee 5/31/21 13 II 50 10.46 _ 10.91 Average..... 48.4 9.19 10.06 The parthenogenetic reproduction of the parallel races was continued over the summer of 1921 without any further obser- vations. The usual care was employed in their feeding. About the middle of August a wild race, C, was introduced. The seventh series of observations was begun with the 169th genera- tion of the A and B races and the roth of the wild race C. Con- siderable variation was found in the number of young produced by successive generations of each race. The difference, however, in the number of young of the B race as compared to the numbers 200 W. V. LAMBERT, W. S. RICE, AND H. C. A. WALKER. of young of A and Cwas marked. The average period of growth and reproduction was 63.6 hours, during which period the average number of young of the A race was 6.86; of the B, 9.35; and of the C, 8.66. The complete tabulation of results may be found in Table VII. TABLE VII. Number of Average No. of No. of Partheno- Date of Young Females Period of Daughter-females. genetic Isolation. Isolated and Growth and Generation. Reproducing. | Reproduction. A B. GC Soe eas | =) TOG ec ts |) of sje 70) 72 hours 9.4 |14.r | 14.9 C—O oss | A | | | B(il0---see: to/ 8/21 Io 66 hours 9.8 IO.I | 10.33 (C=: Cente ae ; A | | Bl lilesssses- Io/11/2r Io 56 hours 6.5 OAL || 7de, (CTC) he en eee | | A | | Bl il4eccseeee 10/22/21 0) 54 hours it Apis || Welyl GAaUSh aaa we bak A Fa Sone 10/25/21 | IO 70 hours 7.6 10.4 8.5 CTO) asaiers oe Averages....... | | nO) 63.6 hours 6.86 | 9.35 | 8.66 | | The three races were then carried on side by side. Since the 145th generation the green food given Race B has been made up for the most part of Euglena, in place of the Chlamydomonas heretofore used. About the same time as the change in the green food the centrifuging of the Polytoma culture was discon- tinued and more drops of the unconcentrated food given instead. After the seventh series of observations, the investigators began to encounter difficulty with the supply of Polyjoma. However, all three races of rotifers continued to reproduce partheno- genetically, although with slightly less than usual vigor in Race A. Nevertheless, notwithstanding all care possible in feeding, isolation and temperature the A race, fed only Polytoma, died out with the 198th generation, probably due to the faulty ae 4 aeest 201 FOOD AND PARTHENOGENETIC REPRODUCTION. ‘moTjONpoAde1 JO 9}e1 ay} Ul eUTpep OU peA\oYs ‘aIp pextur ve uo ydey ‘g vey SvoTeYyM “poul[Iep Ayyenpeas ‘qorp vwojKjog ind ay} uo jdoy SVM YSTYA ‘y oOVY jo uoljonpoides Jo e781 SY, ‘“Spoled esey} SULIMp peur s1oA\ $}S0} OU souls ‘TeoryoyJOdAY v1e g6I 0} QAI puv zz OF I UOMeIOUSS WOT] SOUT] OYL “MOSTIVdUIOD IO} S[AJOJUT ye PeoNporzUT ‘-D Sows PITA\ OY} JO “ToYIO OY} ‘g aoeyY Jo ‘uayorq oy}

Factors Related to the Deueseaiien. of: Littoral Invertebrates. PU Wes eB ae AMDT 205 ie a ates i ‘Pustissep Meneses BY Bee pets "MARINE BIOLOGICAL LABORATORY oe i ‘ ee PRINTED AND ISSUED Bye eee ‘THE NEW ERA PRINTING COMPANY, Ise. LANCASTER, PA: hs eM ~ AGENT FOR GREAT BRITAIN ies 3 WHELDON & WESLEY, Luarep 2 3 ) 2,3 and 4,Arthur Street, New Oxford Street, London, W.P.2 —— Single Numbers, 75 Cents. Per Volume (6 numbers), 53.00 a a = ed \etober 10, 1902, at Lancaster, Pav. as second-class matter under Act of Conaress, of ae I6 1894. ; Editorial Stat vB. G. Conxiin— Princeton University. | >» Grorce a Moore— T he Missouri Botanic Garden. : oT H. Morean— Columbia University. : Ww. M. “Wuezter—Harvard University. E. B. Witson— Columbia University. | ‘managing Editor a FRANK R. Lirtis— The University of Chicage ; ‘ * ; aging Editor, the Nuala of: ones seal 5th t to War cis or Woods Hole, Mass., June I 5th to Sept. I5th. Subscriptions and other _ matter should be addressed. to. the Biological Bulletin, Prince and Lemon Streets, Lancaster, oie Ain roe Ue ana &. Hi fi was are Vol. XLIV. May, 1023. No. 5. BIOLOGICAL BULLETIN Si DIES ING MARINES ECOLOGY IT. SOMES = PHYSICAL FACTORS? RELATED TO): -THE DISTRIBUTION OF LITTORAL INVERTEBRATES. W. C. ALLEE, Ma4RINE BIOLOGICAL LABORATORY AND THE UNIVERSITY OF CHICAGO. PAGE I. Purpose and scope of the investigation. ............-.....-.+2eeeeees 205 TRS IVS PROC Share see pete neraun ate ole: yes tie.rg toAwee) |4le)¢l4i\s SH) eS Sp Sel | eae a || Se 4 Z| <4 Zz n | | i | OPeMED pe rteccormmcnec uss 12 Bua ay | BAW Roe, |) LO | Bir ull Awe | BO | Uo5) Fos Wihart pilings. 22... .-c AN BOS) QD iim |e 2 law || BOO) mw ps | BO | is) Ons exposed) rocksp «sa. 6 | 3.02) 3 | 8| 5.9%, 3 | 22) 22.4) 3 | 9} 8.16) 4 | 13 Rockweedie a16, 4) sscpeneoe 4 | 2.99) 4 | 6 | 6.33, 4 Arts) ab || OG) | Sh3033 3 Se Wii rp ences Aeitin beta, By 2 88s! |r | 2.72) oul ealonkol (Gn lara lye se ier WAU CHESS 3.5 8 6 cd ee oe PON ra 61 ifs) Oo] Blass Gi Biss /O. Be the open water, wharf pilings, exposed rocks, and rockweeds growing on protected rocks. Data from the intertidal associa- tions at the back of mud flats are added for comparisons. The average salinity at low and mid-tides decrease in the order given. The average low tide temperatures increase in the order given from the wharf pilings but these, contrary to expecta- tion, are lower than the open water. The average oxygen tension varies from that found in the open water by amounts which likewise arrange the series in the order given. Thus the wharf pilings gave an average of 4.72 c.c. of oxygen per liter which is 0.49 less than the average found in open water; while the exposed rock association with an observed average of 5.1 c.c. per liter is 0.74 c.c. from the open water conditions. 1 Must range from almost fresh water during rain at low tide to about 3.00 at high tide. 16 238 “WiC, ALLEE, The pH averages the same in the wharf pilings and open water associations. The rockweed association comes next with the association of exposed rocks very near it. The pH values from the exposed rock association is higher than would be expected for the entire coast because only one of the three places in which it was studied was free from the influence of nearby submerged vegetation. Under entirely typical conditions the average pH in this series should likewise arrange the series in the same order as the other factors measured. The rankings given in the table are based upon the extent of variation from conditions known to be characteristic of open water. The sum of the ordinal ranks, a poor enough method of averaging, shows a gradual increase corresponding to the ecolog- ical age of the associations. - ; The range of these four factors (Table XV.) tends to increase with the age of the association. This is not true of each factor taken separately, nor is the increase in total amounts in regular order, in that the rockweed association on this basis precedes the exposed rock association, when it would be expected to follow it. In addition to the absolute range the data from these extremes | show another set of relations not given in averages, that is the relative position of the extremes. In salinity of the open water the lowest record is fairly high and, while the upper extreme of salinity remains approximately constant, there is a general fall in the lower limit as one passes from open water to the older associations. In the temperature series the characteristic change is in the matter of the greater maximum in the older associations. With pH, the young associations have practically no change, the older ones have an increased upper limit due to the action of plants, while the oldest ones show a decided decrease in the minimum on account of the acid-giving muck which is deposited in them. A combination of the rankings based on the departure of average low tide conditions from those known to occur in open water and on amount of variation between observed extremes, places the associations in order of their ecological age, with the exception of the rocks and rockweed associations which are placed together. If in place of an index figure based on amount STUDIES IN MARINE ECOLOGY. 239 of variation of extremes one substitutes the more complex index shown in Table XV., which combines the amount of variation with the location of extremes, the series is arranged by the data given here exactly in order of ecological age. TABLE XV. ERODING SHORE SERIES. Showing range of variations of temperature, oxygen content, salinity, and pH at the stratum most studied in each animal association; all tide stages considered. The last two associations are added for comparative purposes. Ordinal ranking is based on amount of variation and relative location of extremes. Salinity Oxygen Temperature in in ¢c.c. in Degrees pH. Ranking. Per Cent. Liter. Association. oO oO ey wn oO iv Zz o |S y D o S| oo Eels SO Sele |) 2 Sel Ss | ee Sel ee | eu |S El é Béla /é Sel S/ 38] & | Seles Eee Open...... 3.02 3.03| 20 8.0— 3.14 0.12] I | 5.43|/1.50|/1 | 22 | 2.0] 1.5] 8.0+! 0.01] 2 Swe mme || a Wharf pilings..... 2.97 4.06 19 | 8.0 3.13| 0.16} 2 |6.00| 1.94) 2 | 21 | 2.0] 1.5] 8.0 0.0 | I 6.5] 13 2, Exposed rocksas sae. 2.04) 3.62 20 8.0 3.04) 0.10) 3 | 8.56) 4.94) 4 | 23/3.0'3 | 8.3 0.3 | 3.5] 13.5| 26.5] 3 Rockweed .| 2.97) 4.75) 19 8.0 ages 3-02 0.5 | 4 | 8.56) 3.81) 3 | 23 4.0 4 8.3 0.3 | 3-5| 14-5] 29.5) 4 Mayas) 02 2.22. (0.01 19 ea 3.10| 0.88| § | 4:86] 4.86] 4.5| 28] 0.0/5.5 8.2 {2.1 | 5.5| 20.5] 41.5] 5 Melampus .| slight to ? high ee j high tide | tide cond.2 | 6 |high tide! 4.5] ? 5.5, high tide | 5.5] 21.5] 44.5| 6 b. The Depositing Shore Series—The flats series intergrades more closely than the different associations of the rocks. The conditions characteristic of each association are most marked at low tide and these are summarized as averaged data in Table XVI. As might be expected the average salinity at low tide decreases regularly as one goes back on the flats. But this is the only factor considered that shows such a regular relationship. With the others the conditions over the sand bar approach those 1 Hstimated between tides. 2 Estimated for purposes of comparison. ‘“Combined ranking’’ refers to the rankings given in Tables XIV. and XV. The final ranking is based on the preceding column. 240 W. C. ALLEE. of the open water most closely and the conditions in the inter- tidal associations at the back of the flats are normally most extreme, but in the middle regions the plant growth, mainly TABLE XVI. FLAT SERIES. Showing the average salinity, oxygen content, temperature and pH of different animal associations of the flats as found in collections from the bottom, open water excepted, at low tide, except for salinity where both low and mid tides are averaged. Ordinal ranking is based upon variation from conditions prevailing in Open water. Oxygenin | = oD Salinity. c.c. per L, | Temperature. pH gs | dj | eee, Associations. 8 ‘ : ‘ : : ll i 4 re 3 Si ay] S & Md a & 4 = &) hd Ss OS Se OS teen Sil re Ors aire aes Sif |2le|e\eiel 8 |e) a) & leis Ales CA cd z| < Z| <4 g o a CONDE y 0/3 ee Nee eons Ut Ney at ee asset ae || a) |) rae Tel any |) aes Weeetoyallae || at Scoloplos fragilis........ 3 | 3-04] 2 | 3/ 4.23] 3 | 6| 21.3) 2 | 6/8.04 2 | 9 Phascolosoma..........- 6| 3.00] 3 | 4] 6.15] 4 |r2|23.3| 6 | 13 | 8.15) 5 | 18 ISCOLOPIOS GCULUS. J5 7. ©) | 2:07)\) 4a ONNSeSSi Set) 23h 5a FeO Sioa ele) TRU NONE EU Atocsidudshors tite 6:| 2:00| 15/5) \ ON ALSSimany 102200] 3 1|0 8 7eomeianeey SWISGT a ee oe 512.88], Goll t leyel Oley 124) \ 7 ler rege Omles IA AGH EUS E Gla ep OAD AC ©] =| 277 2 Qi OB iN 22205504 ees mee eel grass, affects conditions so that the relationship is irregular. In the summation of ordinal rankings the effect of this confusion is shown. The amount of range, Table XVII., also increases steadily as one passes back on the flats only in the case of the salinity measurements. The oxygen content and pH begin to increase in observed range as soon as the eel grass is encountered in the muddy sand association and reach their maximum range in the tall eel grass of the Scoloplos acutus grounds. The water temper- atures also increase in range but become greatest in the low water of the marginal muck association. The surface in the older associations undoubtedly becomes much warmer when exposed to air and this must be particularly marked in the Melampus association where the surface of the ground may be exposed to the full glare of the afternoon sun with almost no protection from the scant growth of marsh grasses. 1See note at bottom of Table XV. th PS STUDIES IN MARINE ECOLOGY. 241 TABLE XVII. FLAT SERIES. Showing range of variation of temperature, oxygen content, salinity and pH near the surface in the open water and at or near the bottom in other associations. All tide stages considered. Ordinal ranking is based upon amount of variation and relative location of extremes. Salinity in Oxygen in c.c. ore per L. Temperature. | pH. Ranking. ee ore calls s Association. EZ a y 3 3 a g 3 a g s a ee elee a = cs! S S (SI S rQ E\ale| a) 28/5) 8 lel S| 2 lelae| sss OM| & Open..... 3.02 | 3.93 20 8.0 — AoA pW) Te | Sei TOK an || eye PON || Sse Oo || ae |p & |) a Scoloplos fragilis....| 2.99 3.91 20 7.6 3.10| 0.11} 2 4.63] 0.72|2 !24.5) 4.51218.25 |0.65'2) 8 |17 2 Phascolo- SOMMane sae 2.07 4.21 21 Gea 3.10] 0.13} 3 |10.39|6.18| 4 | 27 6.0/3/8.5 |0.81/3]13 | 31 3 Scoloplos acutus..... 2.88) trace 18 7.3 3.02] 0.14|/ 4 | 9.31| 9.5015 | 26.5] 8.5/4 18.45 |1-15|/4)17 |34 | 5 Thyone....| 2.22 B23} L7 : 7.4 3.10| 0.88) 5.5} 5.31] 2.08/3 |28 |11.0/5|8.2 |0.8 | 5 | 18.5] 32.5] 4 Vig) eee oe 2.22 0.0! 19 Fick 3-10! 0.88] 5.5] 4.86} 4.86] 6.5] 28 | 9.0716/8.2 |1I.r |6|24 | 49 | 6 Melampus. | slight to ? high tide high tide| 7 | high tide | 6.5 i Fg) GN Deas) 52s 7 cond. high tide The relative position of the extremes in salinity is much the same as in the eroding shore series. At high tide the salinity is practically constant throughout, but at low tide the lower limit decreases regularly with distance from the open water. The relations of oxygen content, pH, and temperature at the bottom are exactly similar to those of the preceding series except that they are more striking. The ranking in this series, as in Table XV, is eee upon the amount of range combined with the relative location of the extremes. On this basis salinity, temperature, and pH arrange the series in order of the ecological age of the different associa- tions. On the basis of oxygen content, the marginal muck association is much younger than is expected. The combination 1See note at bottom of Table XV. 2 Water temperatures only considered. 242 W. @. ALEEE. of these rankings arranges these associations in order of their age but when combined with the rankings from average condi- tions shown in the preceding table, the marginal muck (Thyone) association is placed slightly earlier in the series than it belongs. Further data would probably adjust this arrangement. At the surface of this series, the range of oxygen, pH and temperature is greatest in the Scoloplos acutus region. The mat of eel grass just at the surface at low tide allows a surface layer of very warm water to be found in the bright sunlight where it is supersaturated with oxygen and: has a correspondingly high pH. The L. litorea, S. lucie, Bugula, C. convexa, B. alternatum, Molgula, and Botryllus, which occupy this region must be very resistant to these extreme conditions. Under the most pro- nounced conditions few animals capable of moving are found at the surface. The differences on the flats level off at high tide. The tide appears to come to the back of the flats over the surface of the more stagnant water which has remained behind during low tide, bringing lower temperature, higher salinity, lower oxygen content and lower pH. Thus two collections from the back of the flats in 4.5 feet of water on a flowing tide showed a specific gravity of 1.018 at the bottom and 1.022 at the surface. Obviously such conditions do not prevail long and by diffusion the gradients disappear. In the long eel grass (S. acutus association), the vertical gradient at low tide in one foot of water was found to be from 3.37 c.c. oxygen per liter to 10.39 at the surface. Three hours later in the same place in three feet of water of the new tide the gradient was from 7.03 at the bottom to 8.3 at the surface. Similarly a pH gradient of from 7.7 at the bottom and 8.7 at the surface became one of 8.0 to 8.1. The condition of the water over the bottom of the flats at high tide tends to become uniform throughout all the associa- tions. Thus at NW. Gutter Flats, collections made under typical high tide conditions showed a range of 0.1 in pH over the flats while at low tide the range at similar stations was over 1.2. V. DISCUSSION. Sumner (1908) in discussing the study of the distribution of bottom living aninials in the Woods Hole region considered STUDIES IN MARINE ECOLOGY. 243 character of bottom, depth, temperature, salinity, purity of water and currents as the important factors in determining distribution. The first of these factors, character of bottom, is characteristically different in the associations of eroding and depositing shores. The former is characterized by the presence of firm places for attachment and difficult burrowing conditions; the latter, by the converse of these conditions. The substratum also serves largely in distinguishing between the different asso- ciations of each series. The influence of depth as a factor in animal distribution in the region covered by these studies is not due to depth as such but to depth as assuring a constant supply of water. This is shown in the rockweed association where animals normally found below low tide level on the wharf pilings may live well above it when protected from drying and from high temperatures by the mat of rockweeds. Temperature serves as a limiting factor for these associations, during the season of the year studied, in the tide pools and more particularly on the flats. There in the Scoloplos acutus associa- tion above the dense eel grass mats at low tide the high tempera- ture (32° C.) that may be reached must serve to kill off the more sensitive sessile animals as it drives the motile ones below the surface layer to the cooler water in the shade of the eel grass. The temperature to which animals may be exposed at low tide increases as one leaves open water conditions in both series. The effect of this high summer temperature of the flats as a factor limiting the geographic distribution in this region will be discussed in Study IV.! of this series. In the associations studied, the salinity regularly increased as one approached open water conditions. The low salinity on the back part of the flats particularly in the Melampus, Mya and Thyone associations must serve as a limiting factor. These are subjected to such extreme ranges of salinity following heavy rains, particularly if these rains come at low tide and where there is some considerable surface drainage, that sensitive animals or animals in a sensitive stage in their life history must needs be killed or driven off. In the locations studied, there was no contamination from 1Jn press in Ecology. 244 W. C. ALLEE. sewage wastes, so this possible factor in distribution may be dismissed. The effects of tidal currents are well illustrated in the gutters and creeks of Hadley Harbor. These protected channels, sup- plied with constantly changing water which differs from open water only by the effect of coming over large tracts of submerged . vegetation, support a wholly different animal life from that present where such currents do not enter. These locations are in the exposed rock or the rock-rockweed stage of development, in place of the Mya or Melampus stage which they would occupy if the currents were absent. In addition to these scouring effects of tidal currents, they have the well known function of oxygen and food carriers. They also eliminate the depth gradients in oxygen and pH found commonly in the stagnant water of the older associations. The relation of oxygen and pH of the sea water to animal distribution has received no attention in the Woods Hole region. Both depend (1) on the supply of offshore water, (2) the amount of photosynthesis being carried on nearby, and (3) the character of the bottom. In the open water these factors depend upon currents and the proximity to vegetation. In the presence of vegetation oxygen is given off and the pH is increased. Muck absorbs oxygen and lowers the pH while sand has no effect on either unless it has been laid bare, when it, as well as rocks under similar conditions, decreases the pH without affecting the oxygen supply. Wave action has the converse effect. From examining data from such regions as the rockweed or eel grass one might be tempted to generalize and say that as the oxygen increases the pH likewise increases and vice versa. Such a conclusion at best holds only in regions of abundant vegetation or of muck, and there the pH changes lag behind changes in oxygen concentration. If the invertebrates are as sensitive to pH variations as Powers found herring to be, this correlation of high hydrogen ion con- centration with the low oxygen regions of the muck must serve to keep free moving animals out of such conditions, Such | action, combined with its greater regularity of distribution and slower fluctuations, makes the pH of the water more important STUDIES IN MARINE ECOLOGY. 245 in such studies as the present than is the distribution of oxygen. Both factors, however, may range widely within a given association. Take for example the sensitive Amarecium associa- tion which flourishes on wharf pilings or on exposed rocks. At the mouth of Northwest Gutter, where the pH stays about 8.0 and the oxygen ranges from 3.72 to 4.95 c.c. per L., the rocks support a typical rock-Amarecium association. Yet the same: association, fully as rich in species, occurs at Gansett on similar rocks, but with much plant life all about and with the pH varying from 8.0 to 8.3 and the oxygen from 5.15 (and probably lower) to 8.56 c.c. per L. It will be noted that the range of conditions either at Northwest Gutter or at Gansett is lower than when the two are combined, so that, while the association can exist in these limits, it is not subjected to these extremes in one locality. The total range in both temperature and salinity is likewise -greater when the two locations are considered together than when either is taken singly. It is probable that a collection of data concerning the conditions under which the different sensitive animals live in all their different localities would, when thrown together, indicate that they could stand widely differing concentrations of all the water factors considered; when, as a matter of fact, they are exposed to relatively slight changes in the location in which they do live. With less sensitive animals the association limits as set out in this paper mean nothing. Pagurus longicarpus apparently roams at will in all of those of the flats and occurs among those of the rocks, carrying with him, willy-nilly, his commensals. The mud snail, Nassa obsoleta, is at present found among all the associa- tions, from the clean sand to the inter-tidal associations, and, according to Dimon, originally dominated the sand also before being driven off by Littorina litorea. In part it is able to do this because it is a resistant animal and in part on account of the fact that it probably becomes accustomed to conditions in a given locality and tends to keep within them. In this regard, one can but express the wish for more studies like that of Dimon. With a series.of such studies at -hand one could draw definite conclu- sions where now, so far as individuals are concerned, he is limited in large part to theorizing. There is so much work necessary in making an ecological 246 Wi ea Ae WDE survey that there is always a lively interest in single factor indices of associations, and at the beginning of the present studies I rather expected to find such an index in the pH relations. While a combination of the average pH, the extent of range and the relative position of the extremes does allow one to place these associations in their natura! order with considerable ‘exactness, and while such data is very suggestive it does not classify these associations with the precision necessary for a successful single factor index. This is emphasized in the rock series of associations, Tables XIV. and XV. The use of the oxygen content of the water is out of the question as such an index of an association. Temperature is an aid but needs confirmation. Salinity, whether average or range, con- sidered with position of extremes, does arrange the different associations in their logical order, and so qualifies for the recom- mendation that was predicted for pH, as the best single index when water conditions alone are considered. The readily deter- mined factors of salinity, temperature and pH taken together give a much stronger index than any one of them alone. If, however, on account of urgent haste, I should be forced to make use of a single criterion to divide the communities of the Woods Hole littoral, I should depend more on observation of the character of the sea bottom than on any other one factor.! This, the most obvious, the longest used, is still the least treacherous single factor index of littoral distribution in this region. It should be used with discretion since a rock well back on the flats supports a different set of animals from one on an exposed point, but the corrections are more obvious and more easily applied. All the data collected in the present investigation agree in supporting the common sense conclusion that animal associations in a region such as this under consideration are not normally limited by any one factor, but by the interaction of several, and when feasible all these should be analyzed and recorded. 1 Shelford (’14) in writing of the suitability of water for fishes concluded that the amount of clean bottom, the amount of carbon dioxide and the amount of hydrogen sulfide, taken together serve as an index of availability of bays and enclosures of the seas for fish life. Longley (’22) in studying the local distribution of Tortugas fishes concludes that the local distribution of many species is determined by the character of the bottom. This holds particularly for what he calls the “‘sand-patch-’ association. 5 vest qnngeia STUDIES IN MARINE ECOLOGY. 247 The ecological age of the different associations has been repeatedly mentioned. This idea clearly and repeatedly stated by such workers as Cowles, Shelford and Adams, is apparently not yet fully understood. Briefly, it means that the Melampus association as it exists at present is an old association which has passed successively through the other stages standing before it in the series. Thus at one time the spot on which a Melampus association is now located was bare sand, which, as it became finely ground and somewhat packed, began to support a Scoloplos fragilis colony much as is found on Blind Gutter Bar at the present time. With accumulation of the organic products of these animals in the absence of a scouring current, other animals came in until the Scoloplos fragilis association began to resemble a Phascolosoma association. As more muck was deposited the older associations were passed one by one until the present old Melampus association resulted. With further deposition the land will be raised above tidal level and the Melampus association will gradually give way to strictly land animals. In this region the muck that accumulates is almost entirely of organic origin (cf. Survey, p. 32) so that the animals themselves have played a considerable part in causing ecological succession to take place. Transitional stages between the different associations are abundant. A particularly noticeable one was studied that be- longs between the Scoloplos acutus and Thyone associations. This is located near the gravel spit at the Uncatena side of the Northwest Gutter passage. There the Scoloplos acutus occur in the muck, but Thyone are absent as yet, although they are found nearby in greater numbers each year. Of the eight comparisons that have been made in the summary tables, four would place this location with the Thyone association, three place it as younger than Phascolosoma association, and the other with Scoloplos acutus where it belongs according to its animal inhab- itants, and according to the average of these physical factors. The wharf pilings present a specialized Amarecium type of association that is obviously younger than the rock-Amarecium community, although closely related to it both in animals present and in physical conditions. It represents more nearly the type of habitat that might be found where large rocks extend up out 248 W. C. ALLEE. of deep inshore water. Being a man-made habitat it is not, strictly speaking, a part of the rock series and is included in that series because it does give approximately the same condition that would be found on the rocky pillars just mentioned. The succession of forms in this association can readily be studied. Glass slides placed under wharfs furnish a convenient method of finding the first pioneers to be expected. Observation of the growth of communities on new non-creosoted pilings and comparison with those on middle aged and old pilings would give the whole story for the wharf pilings, since they will of course never become a Mya association, as a rock-Amarecium association may, or in the same way that its rock pillar proto- type might. In addition to such a study as this and to the studies of the ecology of individual species, in order to describe completely the littoral ecology of this region, studies should be made in the late autumn, early spring and in late spring or early summer. This last is particularly needed to round out our knowledge of the physical and faunistic conditions in these associations during the spring reproductive period. VI. BIBLIOGRAPHY. I. SPECIAL REFERENCES. Pratt, H. S. 716 A Manual of the Common Invertebrate Animals. McClurg, 737 pp. Text figures. Sumner, F. B., Osburn, R., and Cole, L. J. *r11 A Biological Survey of the Waters of Woods Hole and Vicinity. Bull. Bur. Fisheries, 31. Physical and Zodlogical, Pt. I., pp. 1-443. Contains distribution charts and bibliography. A Catalog of Marine Fauna, Pt. Il., pp. 547-795. Faunal bibliography. Verrill, A. E. and Smith, S. I. 7172 The Invertebrate Animals of Vineyard Sound and Adjacent Waters. Rep. U. S. Fish Comm., 1871-72 (1874), pp. 295-778. Plates. (This report may still be obtained from Dr. Verrill.) II. GENERAL REFERENCES. . (Those starred have not been consulted directly.) Adams, C. C. 713. Guide to the Study of Animal Ecology. Macmillan, 183 pp. Bibliography. Allee, W. C. ?1m Seasonal Succession in Old Forest Ponds. Trans. Ill. Acad. Sci., 4, pp. I-6. ’19 ©=Note on Animal Distribution following a Hard Winter. Bu1oL. BULL., 36, Pp. 96-104. 4 : evs : ; “adanaging ‘editor Bos Soe a RANK ae Linue—The University ie Chicago. | 3 Wo Hole. ees I ok £0: Sean de co taebop i ite site matter should be addressed to the Pe Bgl Prince ee ‘Lemon Streets, Lancaster, Pa. ea sith a iy BR ae ore) Vols X BEV. June, 1923. No. 6. BIOLOGICAL BULLETIN CERTAIN EPPECTS OF THE SALTS OF THE HEAVY METALS ON THE FERTILIZATION REAC- TION IN ARBACIA PUNCTULATA. LEIGH HOADLEY, THE HULL ZOOLOGICAL LABORATORY, UNIVERSITY OF CHICAGO. Epler OC CELOM sears ties cuss, ssc sels s-4| aces ales: » GLORCERER ea ol eaiachey c eaeler eters eos ee aie 255 Hes Nonaletertlizationsreaction) in Avbacza: ss... ae ee neice ae 258 WIT IMIS iclavorls "ui ‘ae ede atone heehee LON ME Re mree et Me Ren onc tr ctios. 4 OF aie ye mat Unu ts Pech ie ether 258 Woy LEE OSM ETOUEEL, «Go eS 1 cr CANE RRR eee SCPE che ucla teva tpn rec, CNS 260 Cm ar aOnmvatiousimetales .: cis «nce oa eaenereras ones nee fies 260 De CoOnmpanravuventestsr vane ces <6 ccsia ly oie Rivals ene thane nes ou goesae 271 Ga LUTIMIAA ATG Vege tee sorcerer ais sy hik avy asso be, yn gone ee eSB E eo er enioog mee neon ncteeie cinalel aust 274 Wo LOS CLESSSIOS IRS Sie orci ONS neo URC ROR ORO ae INS CATA se ee eC cnec o ict ea eee 275 WHS CYomavelhitisnorarstsc ata. gece SRStcate RR ORS OG Gaskap Soy oS lateeaa ie e Oac Bey ee 207 I. INTRODUCTION. At the suggestion of Prof. F. R. Lillie in the spring of 1921, I began, the following summer, a study of the effects of the heavy metals on the fertilization reaction in Arbacia punctulata. I desire here to express my appreciation for his many helpful suggestions and his aid in the interpretation of the results. Lillie (21) found that copper salts completely prevent the elevation of fertilization membranes in Arbacia even at such a great dilution as I : 500,000, though the concentration of copper chloride required for the suppression of cleavage is I : 62,500. It is the purpose of the present set of tests to establish the effects of a series of metallic salt solutions on the fertilization reaction and by this means to throw more light on the initial events of | this reaction. The work was done during the summers of 1921 and 1922 at the Marine Biological Laboratory at Woods ’ Hole, Massachusetts. The experiments brought out very clearly that there is a sharp distinction between the initial and the subsequent events 255 17 256 LEIGH HOADLEY. of the fertilization reaction. As soon as the initial part of the reaction is completed, 7.e., as soon as the membrane has raised, the metallic concentration which has formerly been inhibitory in its action ceases to be wholly effective and some of the eggs may develop as far as the motile free-swimming forms in the solution. In the following pages, such a concentration, which inhibits the elevation of membranes, when eggs are inseminated in it, will be termed ‘‘membrane-inhibitory.”” Such inhibition of membrane elevation is complete. If, however, the concentration of the salt is greatly increased, a concentration may be reached at which cleavage in turn is completely inhibited, but here again, the time factor is important. For example, if a given concen- tration is toxic to cleavage for a set of eggs transferred to it five minutes after insemination in normal sea water, it is said to be ‘“‘cleavage-toxic.” It may allow 40 per cent. to 50 per cent. cleavage if eggs are transferred to it 20 minutes after such an insemination, or even 95 per cent. if they are transferred after a 30-minute interval. In such a case, second cleavage will be completely inhibited. When cleavage is used as a criterion of toxicity, the period of exposure to the solution will vary ac- cording to the length of time elapsing between insemination and transfer. If, when eggs are transferred to the solution five minutes after insemination, first cleavage is inhibited, then it follows that eggs transferred at a later time will not perform second cleavage, though they may cleave once. McGuigan (’04), v. Euler and Swanberg ('21), Olsson (21), and others have investigated the effects of the salts of certain of the heavy metals on enzyme action. Mathews (04), working with Fundulus, compares the physiological action of the heavy metals with their solution tensions. It was with the same idea that McGuigan worked on the ferments. He found that, generally speaking, a low solution tension was indicative of great toxicity. There were, however, certain exceptions to the series. v. Euler and Swanberg attribute the poisonous effect of the salts to their property of binding certain groups of the enzyme mole- cule and thus making it ineffective. In the ferment experiments just cited as well as in the experi- ments made by Mathews on Fundulus, the experiments allowed FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 257 the use of distilled water in the solutions, and in consequence, the dissociation of the salts could be accurately measured and precipitation did not trouble the investigators. In the work with Arbacia, on the other hand, it was necessary to use sea- water for all of the final solutions though the stock solutions were made in distilled water. The composition of the sea-water is of such a nature that in many cases there is a precipitation and in no case could the ionization of the salt be accurately determined. The sea-water, as it comes from the tap in the laboratory, has a p.H. of 8.0 and I may say here, that in no case did the concentration of the metallic salt necessary to inhibit membrane formation have a p.H. below 7.0. In some of the solutions there was a visible precipitate, as was most evident in the cases of Zinc, Lanthanum, and Lead. In such solutions, the balance of the salts must be disturbed. Thus, in the case of the alkaline earth metals, and silver, tin, manganese, chromium, and iron, the precipitate was so heavy that after a few preliminary experiments, their use was abandoned. The precipitation of zinc, lanthanum, and lead was not great enough to hide the effect of the salts on the elevation of the membranes but when the concentration was increased for the investigation of cleavage- toxicity, the precipitation was so heavy that it was deemed unwise to try to determine that point for them. When the series was completed, it was found that the work included tests made with the following metals: Gold, copper, zinc, lanthanum, aluminum, platinum, lead, nickel, cadmium, cobalt, and mercury. The results of the experiments show a similar behav or of the metals, save mercury, and those described by Lillie (21) for the chloride of copper. The inhibiting concen- tration varies for the different metals. The behavior of mercuric salts is peculiar and unexplainable at present. Its solutions seem to favor membrane elevation up to a concentration at which the sperm are immediately paralyzed, while at a much greater dilution, cleavage is completely inhibited. This will be treated in more detail in the following pages. Rhubidium and cesium were found to be indifferent up to the isotonic concentration and will not be treated further. 258 LEIGH HOADLEY. Il. THe NORMAL FERTILIZATION REACTION IN Arbacia Punctulata. In order to understand the later discussion, it will be wise here to summarize the phenomena of the normal fertilization reaction in Arbacia. These were described in great detail by Lillie (14), though some of the events had been incompletely described by von Dungern (’02). Fol (’79) described the pene- tration of the spermatozo6n and the elevation of the fertilization membrane in a classical monograph. Before the spermatozo6n comes into contact with the egg itself it is activated and directed toward the egg as described by Lillie (13). The egg, from the time of rupture of the germinal vesicle up to the time of fertilization, or, when fertilization is barred, up to the time of cytolysis, produces an iso-agglutinin (Lillie ’13) or fertilizin (Lillie ’214) which produces a reversible agglutination of the spermatozoa to the egg. Between the time of the agglutination of the spermatozoa to the egg and the beginning of the fertilization reaction, there is a latent period which is very short but variable in Arbacia. This is followed by a very rapid initial reaction which produces a sterilization of the egg against other spermatozoa, described by Just (19) as “a ‘wave of negativity.’”’ The penetration of the spermotozoén occurs immediately, followed by the elevation of the membrane. In a given lot of fresh eggs, the time elapsing before membrane elevation is very variable. At 19.5° C. some eggs appear with membranes fully formed in from 15 to 30 seconds while others require much longer. Elevation of the membrane is too rapid for observation in Arbacia and it seems probable that the variable is due to the latent period. With stale gametes the process may extend over five minutes or more and is often accompanied by polyspermy. III. MerTuops. The experiments described in the following pages are very simple. Eggs were removed from good ripe females by making a circumferential cut around the animal and removing the gonads intact to about 150 c.c. of sea water in a finger bowl. The ovaries were then cut a number of times and the ova allowed FERTILIZATION REACTION IN: ARBACIA PUNCTULATA. 259 to stream forth. In about five minutes, the suspension was strained through cheese cloth to remove the pieces of the ovaries. The eggs were then allowed to settle. When they had settled, the supernatant liquid was poured off and the dish again filled with fresh sea water. This process was repeated three times so that: all traces of the coelomic fluid and the tissue juices were practically eliminated. Eggs were not used unless the normally fertilized control showed above 95 per cent. membrane formation and a good viability through the first cleavage. Males were cut in the same way and placed in a Syracuse glass with the genital pores down. The sperm flows from the pores in a creamy mass, which, when mixed with sea water in the proportion of one drop of dry sperm to 25 c.c. of sea water gives a grayish, opalescent suspension which is described in this paper as a I : 25 sperm suspension. One drop of this mixture when added. to one drop of eggs in 7.5 c.c. of sea water gives 100 per cent. membrane formation with no polyspermy and such an insemination is spoken of as a I : 25 : 7.5. insemination. Concentrated stock solutions were made up as percentage of metallic salt in distilled water by means of the volumetric flask. The solutions used in the experiments were made up by the addition of the stock solution to sea water. Care was taken with those metals which required high concentrations, that the proportion of sea water and stock solution in the mixture used should not alter the osmotic pressure to any appreciable extent. Solutions were not used for more than five days save in the case of AuCl; and PtCl, where the rarity of the metal demanded the use of the one solution. | When an insemination is spoken of as immediate, 7.e., made in the solution itself, 7.5 c.c. of the solution was put into a watch crystal and one drop of eggs and one drop of 1 : 25 sperm suspension added at opposite sides of the dish. The dish was immediately rotated to insure thorough mixing of the gametes. In the viability tests, the eggs were normally inseminated in sea water and one drop of these was then added to 7.5 c.c. of the solution. The dish was then rotated as stated for the immediate inseminations. A number of comparative tests were made in which, owing to 260 LEIGH HOADLEY.’ the size of the experiment set up, it was necessary to use the eggs of a number of females. Unless so stated, each experiment was - made with the eggs and the sperm of but one pair of individuals. IV. EXPERIMENTS. I have chosen to deal with the experimental section of this paper in three parts. In the first I will take up in some detail the behavior of each solution which I used. I have taken the metals up in the order of decreasing toxicity. It will be noted that Mercuric chloride is apparently out of place, but as it gives no membrane inhibiting point in the same sense as the other metallic salts, it has been put at the end. The series of tests described in part 6 was made for the purpose of comparison. The tests show the great similarity in the action of the various salt solutions on the eggs and the sperm, not only when they are unmixéd, but also on insemination. The latter includes both immediate inseminations and viability tests. Under c I have summarized the experimental data. There is also a table there that shows both the membrane-inhibiting concentration and the cleavage-toxic concentration for each of the metals. a. Data on the Various Metallic Salts. 1. Gold Chloride——On July 20, 1922, a I per cent. stock solu- tion of AuCls; was made in distilled water and used for establishing concentrations in sea-water as shown in Table Ia. These were prepared separately for, and at the time of each experiment. A series of tests was then set up to establish the effects of various concentrations on eggs which were inseminated immediately in the solutions. The results are shown in the accompanying Table Ia. It can be seen on examination, that one part of AuCls to 1,500,000 parts of sea-water will completely inhibit membrane elevation. At I : 600,000 and I : 300,000, narrow hyaline zones appear around the eggs after an exposure of about five minutes to the solution, but upon examining the control, it was found to be merely the first indicium of cytolysis in the egg. A series of tests was then made with the purpose of determining what concentration is necessary to inhibit cleavage when the eggs are transferred to the solutions five minutes after insemina- FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 261 TABLE. la. Per Cent. Membranes. | Per Cent. 1st Cleavage. Concentration of AuCl3 in Sea-water. QA. Q B. QA. 2 B. | | Leslee OOOOOO Nr yaa ciniae nares ce I0o I0o 99 roo Tee OFOO OOOO esr eres a sicialee) ov edecievee! 68 20 69 18 Teton 3h OOOLOOOMiehsbeds oistals si seis) bse. sos e's 22 2 20 2 18, SIALLOXYO OVO StS oa neo Uo ee Deen nEee fo) (0) (0) T'S 50 TRA LOYO OKO Oaiarc tho io ancia eNO Re Re (0) fo) (0) (0) I OOO’ OC OOM R ReRMEE or rep trc eo wiaeoana 3 (0) (0) (0) (0) iy 8 OOS OO OM RIOR Oe te katie cranes wists fo) (0) (0) (0) Uninseminated control in sea-water... (0) .o) fo) (0) Inseminated control in sea-water..... 100 100 100 Too Effect of AuCl3 on eggs inseminated therein. July 20, 1922. Temp. 22.5° C. Eggs and sperm fresh. Insemination immediate I : 25: 7.5. tion in normal sea-water. The results for one of these tests is given in Table Ib. Table Id shows the concentration of AuCls toxic to cleavage to be I : 375,000. Comparison with Table Ia brings out the difference between membrane-inhibiting and cleavage-toxic concentrations. TABLE IO. Concentration Per Cent. ; Per Cent. of AuCls in Ist Cleavage. 2d Cleavage. Sea-water. Tem OOKOOOM mire Hike test, eee 90 3 TS GHSOOCOs o piensa a Re Re rc 12 (0) il BT SAO WO er coisa cae NE PERE Ieee os (0) fo) Tin seaialt ys O OO paneer tovslrexelie veivene Site sac oiceysheee iets fe) (0) Inseminated control in sea-water.... 98 97 Test of cleavage-toxicity of AuCls. July 20, ’22. Eggs from @ A. Opened I:34 P.M. Membranes 100 per cent. Eggs transferred to solution five minutes after insemination in normal sea-water. 1:25: 7.5. Spermfresh. Temp. 24° C. In Table Ic can be seen viability tests run with the same lot of eggs at both the cleavage-toxic and the membrane-inhibitory concentrations. This test shows the difference between these two kinds of action. A concentration that is completely inhibi- tory to the initial events has no absolute effect on the subsequent events, while a more concentrated solution will prevent the occurrence of the subsequent events provided the time factor is sufficient. The effect of the exposure time is also shown by this table. The effect on cleavage is cumulative. The higher the concentration of the metallic salt, the shorter need the time of exposure be, to produce complete inhibition. 262 &, LEIGH HOADLEY. TABLE Ic. Per Cent. Per Cent. Per Cent. Membranes. ovale 4 Cell. Time of Transfer after Insemina- ; tion in Normal Sea-water. A B A B. A B Tmmediaten cae. « slscks os) 210 fo) (0) to) fo) (0) fo) ZOVSE CT er darticis eee evens wivis Fine 92 93 20 fo) 12 (0) EOTIIT peptone snare tpersireneciycus: Tete see 95 95 46 (0) I4 (0) Diegales CeN ey ary raat at ieee alee’ 98 08 80 2 65 Ce) HAS OSE ELS AS cs aaa Seem tie IES I0o I0o 84 4 68 Co) Chie te Ie a oe ae BR 100 Too 84 16 80 to) TG Mine MMMM ar oe oe Ser toe, Aas OY 100: 00 92 24 86 fo) £3 O Ma mPa cise ere ey is eae, oe nKOXO) I0o 096 78 95 fe) ‘Inseminated control in GEOAWNUS 6 ogo dco asec onane 100 98 99 Viability tests. Aug.12,’22. 5:27P.M. Temp.21°C. Eggs 9 B. 5:10 P.M. Sperm fresh. Insemination 1:25: 7.5. Solutions: A, I : 1,500,000 AuCls. B,1: 375,000 AuCls. Lilie (21) found that there are three ways in which egg when placed in the membrane-inhibiting concentration of CuCl, can be made to fertilize and produce membranes. Two of these are by means of protective agents. In the first case he used egg sea-water of high agglutinating power. In the second, he used a solution of gelatine or of gum arabic. The third method is by the use of an excess of sperm. A test was made of the effect of increased.sperm concentrations on gold inhib:tion with the results shown in Table Id. A concentration of AuCl; of TABLE Id. Drops I : 25 Per Cent. Per Cent. Sperm. Membranes. Ist Cleavage. I (0) oO 2 (0) (0) 4 (0) fo) 6 4 3 8 20 20 ee 36 34 20 100 92 much polyspermy. Effect of increased sperm concentration. Aug. 7, 1922. 10:38 A.M. Temp. 25°C. Eggs9:50 A.M. Sperm fresh. AuCls I : 750,000. I : 750,000 was used in this experiment. Such a concentration has a p.H. of 8.0 and has no visible effect on the spermatozoa. 2. Copper Chloride—Lillie ('21) investigated the effect of a hie eee eer er < Beste a ae a ae ng rg A age WET ane Whey Se A aL OY & FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 263 copper salts of the fertilization reaction of Arbacia and found that there was a membrane-inhibiting point at I : 500,000 He also determined the cleavage-toxic concentration as I : 62,500. A repetition of this work confirmed the determinations that he made save that for one of the females employed it was found necessary to use a concentration of I : 37,500 CuCl, to completely inhibit cleavage. As has been stated above, the cleavage-toxic concentration varies somewhat, both for the eggs of different females and for the different physiological cond.tions existing in the eggs of the same female. 3. Zinc Chloride.—Zinc, it will be recalled, is one of the metals which formed such a precipitate that a cleavage-toxic concentra- tion could not be determined. In the weaker solution necessary to completely inhibit the formation of membranes, however, the salt could be used with little difficulty. The usual series of experiments for the determination of this point was set up. The result of such an experiment may be found in Table Ila. TABLE Ila. Per Cent. Per Cent. Per Cent. Concentration of ZnCle in Sea-water. Membranes. 2 Cell. 4 Cell. TE Se AACKO(OYOYO)\5 Er ok go ies, SEER Ee 96 46 Present. iS -BXOYO}XOVOKO) 5 Behan arcete oe hemer ack oe Oe 6 5 i‘ IP) GAGE seh iSlolce Hoe crore ose ee eee fo) (0) O° Te MSOF OO ORM mer ure te etnias A Sees (0) (0) Co) EG SUBIR LOXOKOYA “Coan Md! 3 an Oo ese ee RE ee | fo) fo) (0) Inseminated control in sea-water...... 96 96 96 Uninseminated control in sea-water.... (0) Co) fo) Test of membrane-inhibition of ZnCle July 5,’22. 92 C. Eggs and sperm fresh. Insemination I:25:7.5. Exp. 2:30 P.M. Insemination immediate. All membranes narrow. Table IId is a record of the viability of eggs which were in- seminated in normal sea-water and transferred to the ZnCl. solution at intervals as indicated. It will be noted that when good membranes are raised, the toxicity of the membrane- inhibiting concentration on cleavage is not very marked. When narrow membranes have been raised the viability is much poorer. ia Lanthanum Chloride —Gray (’15) found that cerous chloride in a very weak concentration, would aggregate the sperm of Arbacia. He also found that by the addition of a small amount 264 LEIGH HOADLEY. TABLE II. Per Cent. , Time of Transfer after Insemination in Mem- Per Cent. Remarks. Normal Sea-water. branes. 2 Cell. Tmamnediatecs.cs% His Velie oi cae che te ous’ slhtene fo) (0) OVGE Cau ret ener es eiel nie costars ie «eter ee, os aeons 95 63 33 per cent. narrow membranes. Ais OUD Tao cas Pe oro oh aice Cee eR RTE S55 I00 82 Memb. good. DIR Ne eA rete ass Venda wid deh snca ci eens I00 98 HY oe i Mba aene Boe rowed Srey s ca sia sd aoa ve sole emOUe 100 98 s a DNASE dy Bee eke te RIN Se BRO os 100 ~=6|_~—ss 100 es Eg CO oe Dea oN Seer er CR RISES See cg Koyo) 100 oe e TAG us cmeal epee: e 1oRSl circ eh ceinciiah eye's o ates Soe) often eee re 100 | roo ‘ ba OW pelea eee Asae Fic tecasee Se ee eee cella ce eeeMonaaS 100 100 Si s Inseminated control in sea-water...... ’ 100 roo | Viability test. 1:150,000 ZnCle July 5, 1922. Eggs and sperm fresh, Insemination 1:25:7.5. Q@ A. Exp.2:45 P.M. Temp. 20° C. of NaOH, the sperm could be reactivated. Because of this, I made a few additional experiments on the sperm of Arbacia with LaCl; solutions. In the case of the immediate insemination in the lanthanum chloride solution, I found that, at a concentration of N/4,090 or I : 50,000, membranes are not raised. I then made up a thick sperm suspension by adding a number of drops of dry sperm to 7.5 c.c. of the LaCl; solution and observed that the sperm clumped and that their activity was much inhibited. The sperm in this solution were, however, reactivated by the addition of a drop of N/500 NaOH. I then made a similar suspension and added one drop of concentrated egg sea-water. The sperm immediately became very active and gave a very definite agglutination reaction. After the membrane-inhibitory concentration had been estab- lished as I : 50,000, this solution was made up replacing the normal sea-water with concentrated egg sea-water, 7.e., egg sea- water of high agglutinating power. When eggs were inseminated in such a solution, the percentage of membrane formation rose from o per cent. to 64 per cent. to 72 per cent. It may thus be seen that egg sea-water has a great protective action in lan- thanum-inhibition. The viability test of the normally inseminated eggs in I : 50,000 LaCl; solution showed that this concentration is very slightly toxic to cleavage. As will be recalled, a cleavage-toxic concen- tration could not be determined because of the great amount of precipitation. FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 265 5. Aluminium Chloride——The tests with AlCl; were much like those for the metals already described. The results were also much.the same. It was found that a concentration of I : 40,000 (p.H. 7.2) would completely inhibit membrane elevation and that a concentration of 1 : 10,000 (p.H. of less than 6.6) was toxic to cleavage when the eggs were transferred to it after insemination in normal sea-water. As in the case of LaCl;, the sperm were clumped when added to the solution, but when NaOH or egg sea-water was added, this clumping was corrected and the sperm became active. : When the egg sea-water was added, the sperm agglutinated. Table IV. shows the result of TABLE IV. Time of Transfer Per Cent. Per Cent. after Insemination Membranes. 2 Cell. in Normal Sea-Water. GTR TAVEXOHTAN HEA heey Sto aah ene eI IEEE 5 (0) (0) BONSCCTeieteiaciic epi ereeeiate Tile ts: «cit anata sister eae 80 20 IE FUTUC a 6 1 Gc ae Ghee AER ae RES eR PS 97 80 BS OO! TN cath raat ce ee en AR PE 07 92 Ais Sa Ne ERC une) cee iad cc aah Ee Q7 94 6B) oF ot! AA RE aM eee 07 07 Te oT Se cis achea Cicah Syat Ont eee RP ete aa 07 96 2 OMB ET ER Te ee ceisler 97 97 Inseminated control in sea-water....... 97 98 Uninseminated controls. GUSCA=WATCT aa 4 Lina cis coe adele (0) (0) De ANIC Cl eps ie et eS Oe ERE URES sR geR En Eggs clumped. Viability test in 1: 40,000 AICl3;. June 19, 1922. Q@ A. Eggs and sperm fresh. Insemination 1: 25:7.5. Exp.1:00P.M. Temp. 21.5 C. a viability test on eggs inseminated in normal sea-water and later transferred to the membrane-inhibiting concentration of the salt. It will be noticed by an examination of this table, that the membrane-inhibiting concentration is more toxic to the first cleavage than is the case with the other metallic salts. A test of the increase of membrane elevation due to an increase of sperm concentration showed that the percentage was raised from O per cent. to 32 per cent. on increasing the amount of sperm from 1 drop to 15 drops of the I : 25 suspension in imme- diate inseminations. 6. Platinic. Chloride—In the case of no other metallic salt studied is there as sharp a line of demarkation between concen- trations which permit of membrane elevation and those which are absolutely inhibitory, as with PtCls. A stock solution was 266 LEIGH HOADLEY. made from the crystals which was .9 per cent. in strength and | all the concentrations were made up by adding this to sea-water. The test for the membrane-inhibitory concentration shows that at I : 16,666, there is membrane elevation to equal the control, while at I : 11,111, there is a complete inhibition of membrane elevation. The p.H. of the latter concentration is 7.4. In this case each egg may be surrounded by from twelve to sixteen very active sperm. They strike the membrane and remain agglu- tinated to it though they are unable to incite any cortical change whatever. The agglutination of the sperm, by a drop of egg sea-water, in such a PtCl, solution when the suspension is made up directly in the solution, is good. The inhibition comes after the agglutination and before the cortical response of the egg can show itself, z.e., it occurs in the latent period. Not only is there a sharp distinction between concentrations which allow complete membrane elevation and those which are absolutely inhibitory, in the case of this metal, but the membrane- inhibiting concentration and the cleavage-toxic concentration are both more rapidly toxic to the eggs than is the case with the other metals studied. Table Va shows the results of a viability TABLE Va. Time of Transfer after Insemination in Per Cent. Per Cent. Normal Sea-water. Membranes. Ist Cleavage lsgabaalexe bohicras: Rest RR ast otto Sco: chs FO (0) EXONS CHIP Css oh ICRC NERO HERE SCRO ONT asco 88 44 TUE 5 Sec eeersis aliens: 5 URI CLENG SACRE eme 92 50 he ROEM Te seat Ee MRE tents ieh rea, < ‘ohenes 96 62 NAR ay ae een ot Re RODE TE ERAS Che, G8 chs 98 84 5 ily Re ate iri, Ser Ra RIS eee nlee eal y 98 84 i One eee ERE Beh sla ete 98 98 10 EE ee ne Sl MEH Ren Cita od ¢ 98 97 _ Inseminated control in sea-water....... 98 98 Uninseminated control in sea-water..... 0 (0) Viability test in I: 11,11r PtCl.. June 30, 1922. Eggs9g:o00 AM. QA. Sperm fresh. Insemination 1:25: 7.5. Experiment 10:00 A.M. Temp. 21°C. test at the membrane-inhibiting concentration of this salt; Table Vd shows the effect of such a solution on eggs which have been forced to produce membranes by the use of an excess of sperm; Table Vc shows the result of a viability test at the cleavage-toxic concentration. All of these tests show the great poisonous effect of the solution on the egg. The test with the FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 267 TABLE VO. Drops of 1 * 25 Sperm Per Cent. Per Cent. Suspension. Membranes. Ist Cleavage. TCA NeL GB eeS-fkov blot atmo, LGA parE Den eee ee fo) (0) Ghat Shee Na onl! oleae GOO OTTO EEE (a) fo) ZI dem Big & Oldr6 i Bic'o Gis REO eee 2 narrow. fo) SS) a ORO ONA G1. Golo (01 5 JoAnne Cece 410 a 0.5 irregular. TEA 3 SOo Ren IaYGis 0 -Dio SO:cee REO 80 ad fo) AL OPE scp ORR TTT ose seals as ave cas elles 100 op IT Inseminated control in sea-water....... 100 good I0o Uninseminated control in sea-water..... (0) (0) " Effect of increased sperm concentration in immediate insemination in I : 11,111 PtCl. June 30, 1922. Eggs asin Table Va. Exp.10:00 A.M. Temp. 21° C. increased sperm concentration shows the resistance of this solu- tion to enforced insemination as well. Only after the eggs have been left in normal sea-water for four minutes after insemination therein, does the viability in Table Va begin to approach the normal control. At this same concentration, I : II,I1I, while 100 per cent. membrane elevation may be produced by increase of the sperm concentration from one to sixteen drops of the I : 25 suspension, only I per cent. of the eggs cleave and these are irregular. Wein Wee Time of Transfer ‘ after Insemination in Per Cent. Per Cent. Normal Sea-water. 2 Cell. 4 Cell. HO pIMA I pee ge ear see cytes oe wad: vlechaeene eee to) te) 2 OM MMAR eS MAC eo cc 8 a ahalenane (0) fo) OP Men witec neha de iid, oe UG RnR ec ins (0) Co) 1G nacre PCE LN 9208) 8 oo he 2 ; fo) Inseminated control in sea-water....... 96 96 Viability test in 1:1,945 PtCla. July 1, 1922. Eggs 2:00 P.M. Sperm fresh. Insemination 1:25: 7.5. Membranes 100 per cent. Experiment 3 : 30 P.M. Temp. 22° C. A test of the cleavage-toxic point gave that determination as I : 1945. Assoon as the eggs were transferred to such a solution after insemination in normal sea-water, they become crenate. Such a solution allowed no cleavage up to 45 minutes when the eggs were transferred to it at intervals after insemination in normal sea-water. The results of this experiment are given in Table Vc. Cleavage took place in 48 minutes in the controls: 7. Lead Chloride —PbCle, when added to sea-water precipitates very freely. This precipitate is not so heavy, however, that the membrane-inhibiting concentration cannot be determined as I: 4,000 or N/554 (p.H. 7.0.). Some precipitate was present 268 LEIGH HOADLEY. even at this concentration so that the amount of the salt really in solution could not be figured. Sperm will agglutinate readily in such a solution, though the eggs become clumped. 8. Nickel Chloride ——A series of tests, made with NiCle, follow so closely the general behavior of the metallic salts already described that but little discussion is necessary here. It was found the membrane inhibition first becomes evident at about I : 30,000 when the percentage of membranes raised fell from 100 per cent. to 80 per cent. in immediate inseminations, while at I : 5,000, N/324, membrane elevation is completely supressed. Such a solution, when used in a viability test, shows a decided effect on the production of cleavages other than the first. The effect is shown not only by the percentage of cleavage, but also in its regularity. When tests were made to establish cleavage-toxicity, the concentration required was found to be 1 : 600 or N/39 —. Agglu- tination of the sperm is good in I : 5,000 NiCl: when one drop of the egg sea-water is added to 7.5 c.c. of thick sperm suspension made up in the membrane-inhibiting concentration. The sperm are active in such a solution even before the addition of the egg sea-water. 9. Cadmium Chloride——Cadmium differs from the other metals studied in only one respect and there the difference is quantitative and not qualitative. When eggs are placed in the membrane- inhibiting concentration, which is I : 1,333 + or N/122.1, and after various exposure times, are transferred to normal sea-water and inseminated, there is a very marked poison effect after two minutes exposure. Not only does two minutes exposure cause very irregular development, even at the first cleavage, but it tends to inhibit the elevation of membranes. After four minutes, there were 98 per cent. membranes formed and 12 per cent. were narrow. After one hour, all were narrow. When eggs were exposed to the solution for one hour and then insemi- ated, the first cleavage gave 16 per cent. regular forms and 42 per cent. irregular. The effect of this concentration on the sperm is very marked. They agglutinate readily with egg sea-water and can be activated by one drop of NV/500 NaOH or egg sea-water though they are SE ai BS SAE VRE eee ee ae ee ee FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 269 quiescent in the solution alone. The percentage of membranes was raised from O per cent. to 84 per cent. by increasing the sperm concentration at insemination from one drop of I : 25 to 12 drops of the same suspension. The higher sperm concentra- tion of this test resulted in much polyspermy. It would be expected that by increasing the concentration of the CdCl, in solution, a cleavage-toxic point could be determined. I was successful in this; such a concentration is I : 250, or NV 23 —. 10. Cobalt Chloride—When a series of tests was made to establish the concentration of CoCl. necessary to inhibit the elevation of membranes, and the concentration necessary to inhibit completely the formation of the first cleavage in eggs which have been inseminated in normal sea-water and transferred to the solution five minutes after the insemination, it was found that both are much higher that in the other metals w.th which I worked. The membrane-inhibitory point was found to be I : 1373, or N/89.245, while cleavage-toxicity occurred at I : 100 or V/6.5. As was the case with Cadmium, though not to such an extent, the toxicity of even the membrane-inhibiting concen- tration on eggs which had been inseminated in normal sea-water and trans‘erred to it, was very marked. Agglutination is good in the solution. The membrane- inhibiting concentration makes the sperm sluggish but this harmful effect is corrected by the addition, of the NaOH or a drop of the concentrated egg sea-water. 11. Mercuric Chloride—As has already been stated above, there is one metallic salt which is irregular in its behavior. This is Mercuric Chloride. All the other salts with which it has been possible to work have a double effect on the eggs of Arbacia. _ The first effect is at the weaker concentration and is on the phenomenon of membrane elevation; the second effect is con- stantly at a greater concentration and is dependent on the time of exposure of the gametes for the intensity of its action. It modifies materially the viability of the eggs and I have termed it “cleavage-toxicity.”’ The tests made with HgCle show that there was no concentration which would inhibit the elevation of membranes and allow cleavage if eggs were placed in it five minutes after they were inseminated in normal sea-water. 270 LEIGH HOADLEY. TABLE Via. Concentration of Per Cent. Per Cent. HegCls in Sea-water. Membranes. Ist Cleavage. Te TEOOOOOOK. 2 aes fe otes ve vs ee oe I0O 16 Tis ere OOKOOO eau. aie) ose ss ORR ee Melee 100 12 it OOOFOOO I hie cles Scie See ee Meee 100 (0) TH) WAOOFOOO Rf scot gre Oe aude mn Eee 100 (a) I BOO OOO 62:5 oS ac soe) aararegedet eee iqeyo) (0) I DOOOOO' ses een i eee Too fo) Tee ATO OL OW Olen, secon e's) ach) Gl a ROS 52 (a) in 8 FELOOO Ser heise scaetert se er ot ed Te Ren 48 (0) Inseminated control in sea-water..... 100 I0o Uninseminated control in sea-water. . fo) fo) Test with HgCl:. July 11, 1922. 11:20 A.M. Eggs and sperm fresh. Insemination 1:25:7.5. Eggs 9 C. Temp. 20° C. Concentrations greater than I : 300,000 showed cytolysis in 20 minutes. For example, an examination of Table Vla will show that at I : 600,000, HgCl. is toxic to cleavage, while there is 100 per cent. elevation of membranes which are wide and to all appear- ances absolutely normal. In those concentrations in which cleavage does not follow membrane elevation, cytolysis com- mences within an hour for those eggs with membranes. Eggs without membranes are more resistant. It appears from Table VId that at 1: 7,500 HgCl,, membranes are not raised. At first glance it would be thought that this is a direct inversion of the ratio found to hold for the other metals where membrane- inhibition occurs at a lower concentration than cleavage-toxicity. An examination of the effect of such a solution on the sperm shows that they are instantly paralyzed by it and that this paralysis is not reversible, z.e., neither NaOH nor concentrated egg sea-water will activate the sperm in it. This immediately differentiates such membrane-inhibition from that described for the other metals. Lillie (21) notes that the action of HgCle is very different from that of CuCl. He notes that ‘the initial stages are relatively little affected, . . . the susceptibility increases as fertilization progresses’’ (page 140). He also notes that “mer- cury also suppresses the movements of the spermatozoa at great dilution.”” According to his data (Table VII.), a concentration of I : 625,000 completely inhibits cleavage while 1 : 15,625 prevents membrane elevation. He states further, that at the latter concentration, sperm are “paralyzed instantly.” My observations are in accord with the above, save that it was found ; i 3 } p i } ee ee a a ee en FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 271 necessary to use a concentration of 1 : 7,500 HgCle to instantly paralyze sperm and consequently to completely inhibit membrane elevation (Table VI0d). Lillie considered that the action of TABLE VIO. Concentration of Per Cent. Per Cent. HgCls in Sea-water. ° Membranes. tst Cleavage. TNS SROOOR Me eucmeee siciie a7 ia: Poesia aie ele ed ocean 86 fo) aS ONO ODOR Mes teateratentys, 5:75, 6 siontie Sie keer 470 fo) SE ISRO OOMCRRE MG CR Estee A aes sl aus, 6.41 ote mactlesa ee 4o fo) Tee 2 ONO OOP a een a ee ere oiial ees cesta os ees eevee 30 (0) Tae aT OO ORM RII is aie: 5 sa Hit Sia Ah Sone eeee 20 fo) TES ROCOO Ain Gg SiN AO TE ee eee 4 (0) Lee eR 7S OOP are ENA ee sions coke ec lahke nes Some ate (0) (0) Inseminated control in sea-water....... 98 09 Uninseminated control in sea-water..... oO (0) Westwor HsClo july) 25, 1922. 3:04 P/M.) Temp: 26° (© Eggs QE: Eggs and sperm fresh. Insem. 1: 25: 7.5. mercury on membrane-inhibition was like that of copper but at a much higher concentration. I would prefer to consider that the action of HgCle is a poisoning of the sperm, and not the type of inhibition that is present in copper-inhibition. In most of the metallic salt solutions, the effect is to make the membranes narrow as the concentrations approach the strength, necessary for membrane-inhibition; HgCl. has exactly the opposite effect. When eggs are inseminated in solutions which do not paralyze the sperm, membranes elevate immediately and such membranes are extremely wide. The solutions alone will produce membranes on eggs in concentrations greater than I : 150,000 as has also been observed by Lillie ('21). These membranes form much later than those produced by insemination so that it is easy to differentiate between the two. Eggs with membranes cytolyze very rapidly when left in HgCl, solution. It would thus appear that, in comparing the action of the Mercuric salt with those of the metals already described, the only great difference is in the failure of the HgCle to show any concentration which prevents the action between viable egg and sperm and consequently the initial response of the cortex of the egg. | b. Comparative Tests. The preceding part is devoted to individual tests on the behavior of various salts. These tests were made with separate 18 BT 2 LEIGH HOADLEY. lots of eggs, under different conditions, and are therefore not as suitable for comparison as they would be had they all been done simultaneously with one set of eggs. Three sets of experiments were therefore set up for the purpose of comparison. The experiment which appears in Table VIIa, was made to TABLE VIIa. Percentage First Cleavage. Chlorides of: Time of Transfer after Insemination. Au!| Cul} Zn | La | Al | Pt | Pb | Ni | Cd | Co Imimediateren. .asc 1 ceo onan te) Co) (a) fo) (0) (0) (0) fo) o| o BONSCOR caheehcnde Mle siscoace teh amen 29 | 38 | 87| 98| 94] 96| 80} 70} 80} 80 TpAIMM LN poeeterte neice cactee aise eave epee 46 | 64 | 96] 98] 92] 95] 84] 78] 87] o2 Dn eS Fes ie CaP ae eee ere ee 80 | 89 | 98|100!] 97] 96) 92} 90] 98} 94 sh AO Oe ae eee By SRN ER ie 84 | 98 | 100 | 100/100} 98] 98} 92] 100] 98 (33) i eye ane Ee ERE RO ohomeasiavese ee 84 | 99 | 100 | 100/100] 99/100} 98] 100] 99 Les RE Pacha chocier ta core Geen 92 | 99 | 100 | 100 | I00 | 100 | 100 | 100 | 100 | 99 3 OMA Rise Mar ore Acie nek exit 96 | 99 | 100 | 100 | I00 | 100 | 100 | 100 | 100 |} 99 Inseminated control in sea- SW GG Items itha ta-neceuntorpeace eomewenena t= O8) (OOM. LOOW goatee isis ors) sy ape leysiors ae she oreo cee yeneenens Comparative test with membrane inhibitory concentrations. July 15, 1922 Experiment 10:15 A.M. Temperature 21°C. Spermandeggsfresh. Insemina- tion I: 25: 7.5. Membranes formed well, save in immediate inseminations. 1In the above experiment, the results for Au and Cu are interpolated. show the effect of the various solutions on the viability of the same lot of eggs under the same conditions. In each case the concentration used is that of membrane-inhibition. In this table it will be noted that only the figures for the first cleavage are given. In all save the immediate inseminations, membranes were present on a high percentage of the eggs. The table, therefore, emphasizes the great difference between the inhibition of initial and subsequent events, and shows the great similarity of the action of all of the salts. In this table, the figures for AuCl; and CuCl, are interpolated. It was found, however, that both of these membrane-inhibiting concentrations proved more toxic to cleavage than those of the other salts. I have not used HgCl, in this test as it shows no membrane-inhibiting point. The experiment summarized in Table VIIb was performed to determine the time when superficial cytolysis first begins to appear in uninseminated eggs placed in membrane-inhibiting FERTILIZATION REACTION IN ARBACIA PUNCTULATA. concentrations of each of the metallic salts. 273 Readings were made at the end of fifteen minutes, one half hour, and at half- hour intervals thereafter for four and one-half hours. In this TABLE VIIO. Metallic Superficial Cytolysis Salt. Begins. Hours. PUN Teepe ets S28 gs So ay ase, Fo4r a ere SRE RE RR eee es i ETEK Coy, ia, 8 GAN Ee Ne Se tk lool eR ang 3 (CCG aS Aa ae eee OP, LR rec Gh ile pees I ZLSB\ CIO «ocho etc dea PERNT oes Oe, Set phy ha ee ia Iz ILE Class Sic eS eee tee aR ERRMISA Pili ye A HE Tot eal Rh 2 HAI CLs etn ee oa ae EMULE Me Rs Ue Boe No ME au 2 TES rpm es ict scp hanes anciee tore pay Sastnvg See REE EES See 34 IPK Clo} esos enoe ana CMB Ac a Ste Alenia tag 34 INK Clos 3 3) ol Suits eae ei ae Temenos Creston EN 18 Saal a oi ae (CrCl Ge oo sa Eames a a RP AOR ap Nc De oe N82 3 3h Col CI a a5 og. asic a a MER Ns ven arr Ghee a eh elat e ein 4 Controlinisea-waters... 6. isa sc ..0eteme eae eee None in 43 hours. Test of time of cytolysis in membrane-inhibitory concentrations. Eggs, 1 drop to 7.5 c.c. solution. Io: 45 A.M. 7/26/22. Eggs fresh, 2 9s. Temp. 22.5° C. experiment, one drop of the uninseminated eggs was placed in 7.5 c.c. of the membrane-inhibiting concentration of each of the metallic salts. The container, which was a Syracuse watch- glass was then rotated to insure a thorough mixing of the gametes and the solution. TABLE VIIc. Metallic Action Salt. Alone. AuCl 3 ..| Active. CuCle.. oe ZnCle .. ee LaCl; ..| Inactive. Clumped. AICI; ..| Active. Clumped. PtCls...| Active. PbCl2 ..| Active. Some clumping. NiCle...| Active. CdClz..| Quiescent. CoCle..| Sluggish. HgCle. .| Paralyzed. I Drop N/500 I Drop Concentrated Egg NaOH Added. Sea-water Added. Very active. Very active. Agglut. Paralyzed. Clumped. Some activity. Clumped. Very active. Very active. 6c“ 66 sé 66 Fairly active. No change. No change. Effect of membrane-inhibitory concentrations on the sperm. July 18, ’22. Sperm suspension made with dry sperm in 7.5 c.c. of the solution. Sperm from two males. The experiment in VIIc was with the membrane-inhibiting concentration of each of the metals and its effect on the sperm. 274. LEIGH HOADLEY. - In this series, a sperm suspension was made by adding a number of drops of the dry sperm to 7.5 c.c. of the membrane-inhibitory concentration of the metallic salt, and the effect of the solution on the behavior of the sperm was noted. A drop of NaOH (V/500) was then added to one preparation and its effect noted. To another preparation, a drop of concentrated egg sea-water was added. The results may be seen in Tab’e VIIc. c. Summary. In summarizing the results of the foregoing experiments, it may be generally stated, that, with the exception of HgCls, all of the metals studied have three effects on the fertilization period of Arbacia. The first and second occur at a greater dilution than the third and involve membrane elevation. When eggs are inseminated directly in certain solutions, they give no cortical response, 7.e., the effect is immediate and prevents any reaction between the gametes themselves; if, however, eggs are placed in this same solution after cortical discharge has begun but before it is complete, a ‘‘narrow membrane”’ results which may be attributed to an incomplete cortical response. The third involves the subsequent events and requires a higher concentration of the metallic salt. It seems from the data to be cumulative in its action. When the eggs are placed in the solutions five minutes after the insemination in normal sea-water, the time factor amounts to approximately forty minutes, as first cleavage is used as an indicator. If the eggs are transferred at a later time, first cleavage may be only partially inhibited but there will be no second cleavage. That is to say, the time re- quired for the salt solution to produce its toxic effect is, in these experiments, about thirty-five or forty minutes. Table VIII gives a comparison of the concentrations necessary to produce these effects. The order of toxicity of the metallic salts will be seen to be the same for both membrane effects as well as subse- quent events. The general results are in accord with those noted by Lillie in his work on copper. Mercury, as has been observed above, is an exception to this general rule. In each case, as can be seen by reéxamination of Table VIIc, sperm may be caused to agglutinate in the membrane-inhibitory FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 275 TABLE VIII. Membrane-inhibition. Cleavage-toxicity. Metallic : : Salt. N Parts. Normality. Parts. Normality. AuCl3..... I : I,500,000 N/1t51,800 I : 375,000 N/37,912.5 CuCl... I: 500,000 N/ 42,030 Eis) 622500) |) 2Ni/) 85405 ZnCle..... I: 150,000 N/ 10,215 LaCl3..... I: 50,000 N/ 4,090 UNG) I: 40,000 N/ 1,780 I: 10,000 UNI en AAs RtGla. seis: 1g 9) atin ici N/ 936 ie WOVls IN GOR AS 12s saa Tee 4,000 N/ 554 NiCle..... I: 5,000 N/ 324 I 600 INT 3X6) == CdClas.: is I,333-33 | N/ TAD. I 250 IN| OR CoClow an Tas mays N/ 89.245 it 5 Too N/ 6.5 is Clove I : 600,000 N/81,000 The above table shows the series of metals as established for both membrane and cleavage inhibition in Arbacia. With the exception of HgCle they are in the order of increasing concentration. concentrations by the addition of egg sea-water, thus demon- strating that the action is not primarily on the sperm before their interaction with the female gametes. V. DISCUSSION. There are three periods definable in Arbacia from the moment of insemination up to the first cleavage. The first involves the behavior of the spermatozo6én alone and includes its activation, migration to the egg, and its agglutination to the egg. The second involves the reaction of both egg and sperm, and lasts from the end of the latent period to the completion of the cortical activation of the egg. It can, of course, be subdivided. The third is concerned with the rotation of the sperm head within the egg cytoplasm, and the migration of the egg nucleus. In other words, it is the preparation for the union of the germ nuclei and the preparation for cleavage. A latent period comes between the time of the agglutination of the sperm to the egg and the initial events of cortical activation, as described by Lillie (’21). It belongs to the first period described rather than to the second. The differentiation between the first and second periods is very markedly emphasized by the foregoing experiments. When eggs are placed in solutions indicated in the text as membrane- inhibitory, and immediately inseminated, the sperm are seen to 276 LEIGH HOADLEY. agglutinate to the egg. This inhibition, therefore, very evidently occurs between the periods of agglutination and of activation, or during the latent period. It may be said that the egg is _ held at the latent period by the solution. As soon as the agglu- tination is seen to occur, there is a block interposed, and the reaction ceases. At the time when this block is first interposed, there is no great harm done to the eggs. If they are removed to sea-water and inseminated, the sperm penetrate and cortical activation ensues. This seems to show that the action involved affects only the very external parts of the gametes and is easily removed. After an exposure of slightly longer duration, the viability of the eggs becomes poorer, but this I would consider to be due to the penetration of the salt into both the cortex, and, later, the central part of the egg and consequently, to a very different mechanism from that which prevents the elevation of the membrane. There are a number of physical and chemical phenomena which may be responsible for this inhibition of membrane formation. The theory of chemical combination was suggested by Lillie. He remarks (21) the similarity of the effective concentrations found to hold in the ‘“‘cleavage toxicity” of his work and in the enzyme poisoning discussed below. The physical action may be of two kinds. It may be an adsorption phenom- enon resulting in either a change of the electrical charge on the membrane which is present before fertilization or an adsorption by molecular groups. Heesch (’21) finds that the charge on cell membranes of Lycopodium spores, leucocytes, and yeast cells can be reversed by the use of La(NO3)3 solutions. Whether this action is the same as that in membrane-inhibition or not, is an open question. Membrane-inhibition might be due to a change of charge of the egg membrane, or to an adsorption of the metallic ions by some of the complex molecular groups in the egg proto- plasm. Were these groups those of the activable substance, this adsorption would explain the partial activation described above. The first of these would involve an electrical action while the latter would result in a chemical inactivation though not a chemical reaction. Lillie (’21) noted the close similarity between the concentra- _ FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 277 tions of CuCl. required for cleavage-toxicity and that found by v. Euler and Swanberg (20) to be toxic to the action of saccha- rase. McGuigan (’04), in an extensive set of tests of the effect of the metallic salt solutions on the action of diastase, determined the concentrations which inhibit its action for a series of metals. Olsson (’21) also worked on the effects of the various salts of silver and copper on the action of amylase. v. Euler and Swanberg ('21—II) found that the action on saccharase by organic substances was not effected by the temperature. They conclude that there is a binding of the enzyme through the aldehyde group. McGuigan compares the action of the salts to the solution tensions of their ions, though he finds certain irre- gularities in the series. His results for diastase and those of v. Euler and Swanberg for saccharase are in very close agreement. The data concerning membrane-inhibition in Arbacia do not agree with the results noted by the above authors for enzyme- poisoning. While chemical activity could explain the inhibition effect satisfactorily, it would be difficult to conceive an immediate correction of this, effect if the inhibiting factor were removed. As has already been noted, when the eggs are removed to sea- water from a solution in which inhibition is complete, if the exposure has not been too long, there is an immediate recovery and the eggs will produce membranes on insemination in the normal manner. The present paper emphasizes the distinction to be made between the initial events in fertilization as they are indicated by the cortical activation, and the subsequent events which lead to the period of cleavage. VI. CONCLUSIONS. 1. The chlorides of the following heavy metals inhibit mem- brane formation in Arbacia in the following order and concen- trations ;—Au—W/151,800, Cu—N/42,030, Zn—WN/10,215, La— N/4,090, Al—WN/1,780, Pt—WN/936, Pb—WN/554, Ni—WN/324, Cd—WN/122.1, Co—WN/89.2. The effect of these concentrations upon the cortical response of the egg is immediate. Eggs inseminated I : 25 : 7.5 in these solutions immediately become bombarded by many sperm, no one of which is able to activate - the cortex of the egg at the above concentration. 278 LEIGH HOADLEY. 2. In contrast to the above, but showing the same general order, are the concentrations of the same metallic salts which are toxic to cleavage. These salts, with their toxic concentra- tions are;—Au—WN/75,825, Cu—WN/8,405, Al—N/445, Pt— N/163.87, Ni—N/39 —, Cd—N/23 —, Co—N/6.5. There was so much precipitation in the case of Zn, La, and Pb, that the concentration of these salts, toxic to cleavage could not be determined. . Mercuric chloride proved toxic to cleavage at N/81,000, while seeming to favor membrane elevation. At a concentra- tion inhibiting membrane elevation, sperm were paralyzed immediately. These concentrations vary slightly for different batches of eggs and show the influence of a time factor, thus giving an interesting contrast to membrane-inhibition, the concentrations of which are more constant and immediate in their action. These concentrations are greater in every case except Hg, than the membrane-inhibiting concentration of the same metal. Cleavage toxicity is a progressive, or a cumulative action. 3. These solutions are not immediately harmful to the gam- etes, but prolonged exposure is injurious to the eggs as can be shown by two sets of experiments. (a) When the female gametes are exposed to the solutions for various periods of time and are then inseminated in fresh sea- water, the viability always stands in inverse ratio to the length of exposure. (b) When the inseminated eggs are transferred to the solutions at intervals after insemination in normal sea-water, the length of exposure to the solution bears an inverse ratio to the viability of the eggs. It will be seen in both (a) and (6) that the time factor is the important one. The action here is again cumulative. 4. With the exception of mercuric chloride, any harmful effect of these solutions on the sperm is corrected by the addition of a drop of egg sea-water. LITERATURE. I. von Dungern, E. ‘02 Neue Versuche zur Physiologie der Befruchtung. Zeitschrift fiir all- gemeine Physiologie, Bd. I. ——9 FERTILIZATION REACTION IN ARBACIA PUNCTULATA. 279 2. v. Euler und Swanberg. ’20 Ueber Giftwirkung bei Enzymreaction. I. Inaktivierung der Saccharase durch Schwermetalle. Fermentfor- schung, III. Jahrgang, pp. 330-393. II. Inactivierung der Saccharase durch organische Stoffe. Ferment- forschung, IV. Jahrgang, pp. 29-63. III. Ueber den Einfluss von Kupffersulfat auf die Autolyse der Hefe. Fermentforschung, IV. Pp. 90-96. IV. Electrometrische Messungen ueber die Bindung des Silbers und des Kupffers an Saccharase und an andere organische Verbin- dungen. Fermentforschung, IV. Jahrgang, pp. 142-183. ’21 Ueber die Regeneration inactivierte Saccharase durch Dialyse. Zeit- schrift fiir Physiologische Chemie. Bd. 114, page 137. 3. Fol, H. %79 Researches sur la Fécondation et le commencement de 1’Hénogénie chez divers animaux. Genéve Soc. Phys. mem., X XVI, pp. 89-397. 4. Gray, J. *r5 Note on the Relation of Spermatozoa to Electrolytes and,its bearing on the Problem of Fertilization. Quart. Jour. Micro. Sci., N. S., LXI., ‘pp. I19—126. 5. Heesch, K. 22x Untersuchungen ueber die Umladbarkeit von Zellen, Zellbestandteilen, und Membranen. Arch. f.d. Gesamt. Phys. Bd. 190, pp. 198-212. 6. Just, E. E. "19 ©The Fertilization Reaction in Echinarachnius parma. I. Cortical Re- : sponse of the Egg to Insemination. BioL. BULL., Vol. 36. p: 1: 7. Lillie, F. R. ’r3 Studies of Fertilization. V. The Behavior of Nereis and Arbacia with Special Reference to Egg-Extractives. Journal of Exp. Zodl., Vol. 14, PP- 515-574. *t4 Studies of Fertilization. WI. The Mechanism of Fertilization in Arbacia. Journal of Exp. Zodél., Vol. 16, pp. 523-590. ‘19 Problems of Fertilization. xii + 278 pp. The University of Chicago Press. 22x Studies of Fertilization X. The Effects of Copper Salts on the Fertili- zation Reaction in Arbacia and a Comparison of Mercury Effects. BIOL. BULL., Vol. XLI., pp. 125-143. 8. McGuigan, H. ’04 The Relation between the Decomposition Tension of Salts and their Antifermentation Properties. The American Journ. of Physiology, Vol. X., pp. 444-451. 9. Mathews, A. P. ’04 The Relation between Solution Tension, Atomic Volume, and the Physio- logical action of the Elements. The American Journ. of Physiology, Vol. X., No. 6. ro. Olsson, U. ’21 Ueber Vergiftung der Amylase durch Schwermetalle und organische Stoffe. Zeitsch. f. Physiol. Chem., Bd. 114, pp. 51-71. Dae BEHAVIOR OF CALCIUM PHOSPHATE AND CAr— CIUM CARBONATE (BONE SALTS) PRECIPITATED IN VARIOUS MEDIAS Witte AP PLicAlii@ONS i@ BONE FORMATION. JAMES CRAWFORD WATT, ANATOMICAL LABORATORY, UNIVERSITY OF TORONTO. WitH THIRTY-EIGHT FIGURES IN THREE PLATES. CoNnTENTS. PAGE AEH O GUE ELON es yay d sca, acgalelesonens. rs ena Malte GE REPRE TEM Teepe neirenes erst Neyo stiaviepabe rename Cae 280 Part 1. Microscopic Study of Precipitation : MD @CHMITGUAE |) alsvsiz ace Siete Bonide foie, Meotapieite col ever eaamedtceier snore ear wayierta: De aduate teen eraser 282 Reaction, in) aqueous: Solutions vm act teem sea etoile 283 Precipitation, of calcium) sphosphateeeen nee eereiiee ceili leis 283 Precipitation of calcium carbonate............ Nc nich Soo Rea 284 PrecipitatonentetcollotdallsolimtiOnsse meer tern ein ieltie titer ate 284 Precipitation On scaleliim sp lOSphatcemer ieee ei rleieierstne rier ieee 285 Precipitation yor y calcein can bOtlate aE eee rei eee eerie 285 Influence of other agents on the precipitate...................--2-- 288 Inflience: Of “diferent ONS: ee oq ose east ecaieiema eiitensiiel = © eye) a) si ete) etroweiepenetezetere 290 Precipitationsin sbloodaserumpa av eee eRe tio ection 291 Precipitation ime ny Aline weak illas CMe xia © ieee tnelsnelel nile statele ic inetener 291 Influence of hydrogen ion concentration....................s+eeee-> 292 IDyicieayierere(or One Soler oo onaagonadec ons eougoanbaddocGoooudod 205 Number on tonmseassum ede bya precipitates ae aie -lel sr ieicns iene nererenckaices 207 Part 2. Examination of Bone: Mechnique™ and anatertall\c ecg eerie oii atelier terete erent er 208 Resultsof examination Of sbOnessereieiiee oilers eerie Be cemcrd hs 300 Part 3. Discussion: Theories regarding deposit of calcium salts...............2.+2-000- 301 Reversibility, Ofmceal cium: reactor pee eieir teers eleiee line ieleier acne 305 Sbicabiotsh per ernie ce InEeR Mor binig.d oo ronicicnan Ao om te oma doe-L 0.0.9 00. Sic 307 INTRODUCTION. There are at present three main theories presented in explanation of the manner in which the matrix of bone becomes impregnated with the two inorganic salts, calcium phosphate and calcium car- bonate. These views may be briefly summarized as follows: 280 CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 281 1. That the salts are deposited in the matrix by a precipitation in situ from the interaction of soluble salts in the blood and the tissue. 2. That the salts are excreted, or secreted, either with the matrix, or into the matrix, by the bone cells. 3. That a complex combination salt known as calcium carbono- phosphate, carried in solution in the blood, is thrown out of solution in the bony matrix by a change in the carbon dioxide content of the tissue, and after precipitation is finally converted into the two components, calcium carbonate and neutral calcium phosphate, in the exact proportions found in bone. The weight of evidence is at present in favor of the third view, especially supported by the work of Barillé and of Wells, but it seems to me that fully conclusive proof that this is the process has not yet been brought forward. The precipitation theory, however, offers a very simple and plausible explanation, and so it was determined to investigate the behavior of bone salts on precipitation and see if any light could be shed on the actual manner of impregnation of the bony matrix. An exact knowledge of the way in which the bone salts are added would be a great aid, indeed, to our understanding of the growth of bone, repair of fractures, changes in rickets and osteomalacia, and other kindred conditions. It would also throw light on the formation of calcareous patches in scar tissue and in arterioscle- rosis, for Wells has shown that calcification and ossification are quite similar processes, and the same two calcium salts are present in both cases in exactly the same proportions. This research work, as finally carried out, involved the micro- scopic study of the reactions whereby calcium carbonate and cal- cium phosphate were precipitated, both in separate solutions and also in the same solution, first in various aqueous media and then in certain colloidal ones. This was followed by a thorough exam- ination of the unchanged matrix of many different bones in the mouse, the frog, the guinea pig, the dog, and human fcetuses of various ages. “For experimental investigation of the artificial process will 282 JAMES CRAWFORD WATT. furnish the best clue to a precise and certain knowledge of the natural one, by showing more clearly how much is due to physical agency ’—Rainey. PART 1. MICROSCOPIC STUDY OF PRECIPITATION. TECHNIQUE. For simple examination of a reaction, a glass slide measuring 38 x 75 mm. was taken and on it were outlined two squares with sides of 20 mm. painted in melted paraffin, so that the wax formed the sides of a shallow cell. Into a cell were placed a small amount of the two reacting solutions to give the desired precipitate, and then a cover glass 22 mm. square was dropped on, supported by — the wax cell wall. If it was desired to keep the specimen, melted wax was then painted all around the edges, and overlapping on to the cover glass. The wax was then coated with thick shellac, which in its turn overlapped slightly on to the glass. The cell was thus permanently sealed. Two cells could be made on one slide and the same reaction could be compared in distilled water and in a colloidal solution side by side. It was found desirable to examine not only the precipitate formed immediately at the line of contact of the two reacting solutions, but also what was formed later, in more remote parts of the solution, during diffusion, and as the concentration of the dissolved salts became less and less with increasing precipitation. To do this required larger cells, and it was desirable to have a fairly uniform thickness of cell, which was accomplished as follows: On a glass slide 38 x 75 mm: in size two thin parallel strips of mica dipped in melted pataffin were so placed as to support the long edges of a glass cover slip measuring 24 x 50 mm., which was placed upon them. These long edges were now thoroughly painted over with melted paraffin, which on hardening cemented the cover slip to the slide. A cell measuring from 0.1 mm. to 0.2 mm. deep was thus made, still open at both ends. By means of a fine glass pipette one of the reacting solutions, M/1o calcium chloride, was now introduced at one end until the cell was nearly two thirds filled. This end was then carefully dried and sealed with paraffin. CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 283 At the still open end was now introduced the other solution— for instance, 17/4 sodium carbonate, or M/4 triple sodium phos- phate—and the last border of the cell was then sealed and the slide labeled. All borders were then painted with thick shellac, which was allowed to slightly overlap the wax on to the glass. To obtain a series of cells of uniform depth, a sheet of mica four inches by two inches in size was split until the required thick- ness was obtained, equal to that of a number two cover glass. It was cut across with scissors into two-inch strips about two milli- meters wide. One sheet thus prepared will cut into approximately forty or fifty strips. - Cells filled with warm solutions sometimes showed broken cover | glasses on cooling. Also in cells completely filled with solutions a slight amount of expansion due to change of temperature was sufficient to cause leakage by loosening the wax from the slide. And so to provide for permanency of the specimens all cells were so filled as to leave an air space to provide for contraction or ex- pansion without undue strain, and were filled with solutions at room temperature, 20°—21° C. Pure distilled water was used in all experiments, all solutions were filtered to make them perfectly clear, and chemically pure reagents were employed in every case. REACTION IN AQUEOUS SOLUTION. Precipitation of Calcium, Phosphate. Calcium phosphate was obtained by the interaction of M/1o solution of calcium chloride and M/4 solution of triple sodium phosphate in water. This strength of reagent gave plenty of pre- cipitate, without it being so dense as to interfere with proper micro- scopic examination. As the two solutions were of different con- centration, there was a different rate of diffusion on the two sides of the line of contact, with corresponding differences in rate and amount of precipitation. Immediately that contact occurs between the two solutions a milky cloud appears, which grows rapidly in size. For a moment, although the cloud is visible to the naked eye, the solution seen microscopically is still clear, but an instant later it becomes vio- 284. JAMES CRAWFORD WATT. lently agitated, and the agitation becomes quickly defined as being caused by very minute particles, which are in very rapid motion, both oscillatory and translatory. These particles, appearing only as small dots even under the highest magnification, grow in size, losing their motion as they increase in bulk, until finally they form the fine amorphous granules which gather in clumps, and masses, and fine films, to form the precipitate. Precipitation of Calcium Carbonate. Calcium carbonate was precipitated by the interaction of a M@/10 solution of calcium chloride with a M/4 solution of sodium car- bonate. The process here resembles that described above for cal- cium phosphate in the form of myriads of fine particles in rapid Brownian movement. All movement finally ceases as these parti- cles agglutinate into large clumps, and then a remarkable change occurs, for the granular masses just formed fade from the sight, dissolving again into the solution, while scattered here and there larger, rapidly growing particles occur, which become defined as crystals, angular in outline, or as small spherical bodies, known as spherules, which are later converted into crystals as they increase in size (Figs. 1, 2, 3, and 25). Two or three hours after the formation the precipitate is all crystalline. It is very rare to see a spherule persisting in pure aqueous solutions. The appearance of precipitates of calcium carbonate in the form of spherules has been frequently noted before. As long ago as 1839 it was described by Link, who, in accordance with this phe- nomenon, thought that crystals at their first origin were fluid and only later became hard and angular, as solidification occurred. PRECIPITATION IN COLLOIDAL SOLUTIONS. The colloids employed were gelatin and egg albumen. Various concentrations of gelatin were experimented with, rang- ing from an 8 per cent. solution, which solidified quickly, to a I per cent. solution, which remained fluid for some time and then formed a soft jelly. As the various concentrations gave similar results the weaker ones were used as a matter of routine, being easier to handle. CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 285 For albuminous solutions Merck’s powdered soluble egg albumen was employed, water being added to make the approximate normal proportions given by Lillie for fresh albumen in the hen’s egg, namely, albumen 12 per cent., water 88 per cent. by weight. In 1857 Rainey published the first of a series of papers on the precipitation of carbonate and phosphate of calcium in colloids, using gelatin, albumen, and gum arabic. His best results were obtained with gum arabic, and he pointed out that the solutions of the gum holding the two salts reacting to give the precipitate should be of very different densities, so that diffusion and the consequent reaction would occur slowly and gradually. I have found, how- ever, with albumen and gelatin, that if the reacting salts differ suf- ficiently in concentration, the concentration of the colloidal solution need not differ in order to secure good reactions. And so in many series of experiments I per cent. gelatin was used: throughout, although in others this concentration was used only with calcium chloride, and a 4 per cent. solution with sodium carbonate or phos- phate. Egg albumen was used always in the same concentration. Precipitation of Calcwuwm Phosphate tn Colloids. The employment of a colloidal solution seems to have little or no influence on the precipitation of calcium phosphate. The de- posit is quite similar to that in aqueous solutions, being fine, gran- ular, and amorphous (Fig. 36), and aggregating to form thin gelatinous veils and clouds. Results in egg albumen were similar to those in gelatin. These findings are in agreement with the work of Rainey, Harting, Biedermann, and others. Precipitation of Calcium Carbonate in Colloids. The reaction here is very characteristic. Calcium carbonate formed in the presence of colloids has a tendency to separate out and persist in the form of the spherules mentioned previously, which have been named calcospherites. The first effect of the colloid is to very much emphasize and prolong the stage in which the particles are small and show active Brownian movement, apparently by preventing their early fusion into larger masses. In a short time, however, larger particles are 286 JAMES CRAWFORD WATT. seen which grow partly at the expense of the smaller ones, which redissolve as described for aqueous solutions. But there are many small particles which apparently do not dissolve, but pass on di- rectly over to the larger form of deposit, by forming a nucleus for the laying down of additional material from the clear solution. The precipitate exhibits two separate forms. One of these is distinctly crystalline (Figs. 8, 10, 12, 13) and is found most densely at the primary line of contact of the reacting solutions where de- posit was very rapid. Even here the colloid shows its influence, for the crystals are not perfect, but are deformed in various ways, and exhibit rounded angles and suppression of their typical shape. The second form of deposit is in the shape of very perfect small spheres (Figs. 7-12, 29, 30, 31), which vary greatly in their trans- parency, markings, and size according to the solution in which they are formed. Rainey obtained spherules in gum arabic only after the lapse of an hour or more. At first a faint nebulosity was seen, lasting in thin solutions of gum for one hour, in thick solutions for over a week, after which spherules occurred. In my work I have been able to identify these bodies in less than ten minutes, and in some solutions to see them attain a size of 30 in one hour. They grow rapidly and attain practically their full size in forty-eight hours, ranging then according to the solution in which they lie, from a size of Iu up to 120p. In Rainey’s case they grew gradually for - several months and ranged in size from 2p to 200p. I had one single example of a pear-shaped body which measured 120 p x 300 p. Evidently gum arabic exhibits a more powerful colloidal influ- ence on the precipitate, but gelatin or albumen will give excellent results in a very much shorter time, which is a valuable consid- eration. Brownian movement in parts of some slides is still active after the lapse of a year, due to the colloid retaining some of the pre- cipitate in its original, extremely small, granular form, preventing coalescence to form the larger particles seen in other areas where precipitation was denser and more rapid. The process of coalescence of the small particles can be observed in favorable cases. Very small spherules showing active move- AS, CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 287 ment travel about in the solution, colliding with and working past others. Suddenly two of these, instead of passing, will be drawn together, as though pulled by a strong attractive force, and will merge completely in one large spherule, which in its turn may absorb small ones. At these moments of coalescence the spherules give the observer the impression of being fluid, flowing together like two drops of water, and lead one to believe that Link was correct in interpreting these small spherules as liquid crystals. As these bodies become larger they lose their movement, settle down, and become of glassy hardness. There is also a later growth by addition of material from the surrounding clear solution. Richards and Archibald have shown that the particles-of pre- cipitates at their first appearance seem to have a rounded outline to the eye even under high magnification, but by photographs, how- ever, they have demonstrated that these particles are real crystals of typical shape. This condition is true for some particles of calcium carbonate formed in aqueous solutions, but in colloids a great deal of the deposit is in the form of particles which are round, and persist in this shape as they grow. There are other particles also, which clearly are not round, and give the crystalline form of deposit which occurs simultaneously with the spherules. The spherules formed in gelatin (Figs. 13 and 31) are very perfect, with a smooth surface, and clear, transparent, glassy ap- pearance. Those formed in albumen (Figs. 26 and 29) often show a slightly irregular periphery and give indications of their internal structure. They appear to be built up of concentric layers, and often exhibit an amorphous center. Extending from this center to the periphery is also a faint, but quite distinct, radial striation in certain cases. These appearances have already been well described and illustrated by Rainey, Harting, and Biedermann. It ought to be noted here that Harting has been erroneously credited by many writers with the discovery, in 1872, of the calco- spherites formed in colloidal solutions. Rainey described and gave accurate illustrations of these bodies in 1857, and had observed them as early as 1849, and so deserves the credit of this discovery. Combined precipitates of calcium carbonate and phosphate (Fig. 36) show each salt acting quite independently, the phosphate occur- 19 288 JAMES CRAWFORD WATT. ring in its usual amorphous form, the carbonate in typical crystals and spherules. Similar conditions were found by Rainey in gum ‘arabic. The colloids gelatin and albumen were chosen as being artificial media easily prepared which might be looked on as approximately comparable respectively to the bony matrix and to the blood serum ‘or to cellular tissues. Reactions in gelatin might give a clue to ‘those in bone, and similarly reactions in albumen might indicate processes of calcification in other tissues. After studying the re- actions in these simple media, they were further modified by the addition of various substances which are either normally or patho- logically found in the blood or tissues, to see what effect these ‘special substances would exert. Further experiments were also carried out to study the effect of ‘the hydrogen-ion concentration of the colloid. To complete this stage of the work reactions were followed in cartilage extract and in blood serum. Then specimens of bone from various animals were examined to see if there was any correlation between the ‘structure of their matrix and the results obtained experimentally in the various media. -¥ THE INFLUENCE OF OTHER AGENTS ON THE PRECIPITATE. The substance whose influence was to be investigated was added in equal concentration both to the sodium carbonate or phosphate, ‘and to the calcium chloride solutions which were to react to form the precipitate. The new agent was thus balanced so that its action could not be one sided. The agents used and their concentration in solution were: ‘-M/to sodium chloride, same proportion as in human blood serum. 0.1 per cent. dextrose, same proportion as in human blood serum (Gradwohl). 0.05 per cent. urea, same proportion as in human blood serum. O.I per cent. lecithin. Blood serum contains 0.7 per cent. (Wells). 1.0 per cent. acetone. Simulating conditions in acidosis. 0.5 per cent. ethyl alcohol. The effects produced on the precipitation of calcium carbonate were quite definite for some of these substances, and in some in- ae a CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 289 stances very remarkable changes of form resulted. The influence on calcium phosphate was much less striking, as this deposit re- mained in the same granular, amorphous form in all cases, and the results were mainly a slight difference in the average size of granules, and also differences in the speed and completeness of precipitation similar to those which will be noted in detail below for calcium carbonate, to which latter deposit alone the succeeding paragraphs relate. Addition of Sodium Chlorsde. The main effect noted on adding sodium chloride was the speed- ing up of the reaction. Brownian movement ceased sooner, crys- tallization was more rapid, and in most cases the main portion of the precipitate consisted of smaller crystals and spherules, and in the colloids the crystals were of very poor shape. Addition of Dextrose. The influence of this compound was slight and seemed to consist mostly in added transparency of the crystals and spherules. There was also slight modification in shape. Addition of Urea. Even in aqueous solutions some spherules were formed when urea was present. The effect in colloidal solution (Figs. 30 and 35) was to favor increase in size of the spherules, with a marked tendency toward fusion, so that many double forms and dumbbell- shaped bodies were seen. Urea thus shows an influence quite the opposite of sodium chloride upon the precipitate. It evidently favors growth of this deposit and suggests the idea that the pres- ence of urea in blood and in urine may aid in the deposit of calcium salts in the calcareous incrustation of sclerotic arteries and the growth of renal and vesical calculi. Addstion of Lecithin. Lecithin, even in small amounts, 0.1 gm. in 100 c.c. water gives a turbid solution even after filtering, but microscopically the solu- tion appears clear, so that it is evidently a colloidal one. The lecithin has a remarkable influence on the precipitation of 290 JAMES CRAWFORD WATT. calcium carbonate. Even in aqueous solution (Figs. 4, 5, 6) spherules occur in great numbers, and many of these are of a large size not obtainable in other solutions. There is also a distinct tendency toward fusion of spherules in contact with each other, so that double and multiple forms are numerous. Beautiful rosettes (Fig. 5) were present, due to the adhesion of many oval-shaped -bodies. In gelatin and albumen (Figs. 10 and 11) the same phenomena occurred, producing magnificent rosettes and very large spherules measuring 120» in diameter. Lecithin thus has a very typical colloidal action on the precipitate, favoring the spherical forms of deposit. This compound is found in blood in seven times the strength it was employed here, and is also found in bone and other tissues, so that its action on a pre- cipitation of bone salts, if such occurs, ought to be appreciable. This point will be discussed later. Addition of Acetone. The presence of acetone (Fig. 12) hastens the process of pre- cipitation, making it occur very rapidly, and also makes the initial deposit at the moment of contact much denser and more complete. This point is rather suggestive. It leads to the question as to whether or not conditions of acidosis such as accompany serious cases of nephritis and diabetes have an influence favoring the prog- ress of calcification in the arteriosclerosis so often found in these cases. Addition of Ethyl Alcohol. The alcohol had no appreciable effect upon precipitation except to add to the optical clearness of some preparations. Tue INFLUENCE OF DIFFERENT IONS. The calcium carbonate was precipitated in separate cells by the carbonates of sodium (Figs. 1 and 4), potassium (Figs. 2 and 5), and ammonium (Figs. 3 and 6), so that the solutions contained the three last-named ions as the only variable, all other factors being equal. In aqueous solution the crystal form of the calcium car- bonate in the three cases was noticeably different. In colloidal CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 291 solutions the spherules seen were very different in their appear- ances in each set of slides. The most noteworthy change was seen where the ammonium salt was used, the deposit being much coarser, with rougher, more irregular surfaces, and as a rule much greater in size and with more tendency toward fusion and toward clumping into masses. The whole series for each ion, including those slides with the six accessory agents described above, showed similar tendencies to the plain, unmodified reaction. Each series consisted of 21 slides, consisting of 7 in water, 7 in gelatin, and 7 in albumen. PRECIPITATION IN BLoop SERUM. Clear serum was pipetted off from test tubes of dog’s blood allowed to stand for some hours in a refrigerator. The precipita- tion of calcium phosphate in this serum showed the usual granular amorphous character. Calcium carbonate separated out almost en- tirely in the form of spherules, many of which were quite large. The process of precipitation in blood serum was very complete, covered practically the whole area of the slide, and seemed to be greater than usual in amount. This appearance suggested the thought that the sodium carbonate and phosphate of the blood aided the reaction, making precipitation greater in amount, but further consideration made this seem unlikely, as the usual amount of car- bonate or phosphate was added to the blood, and this alone was sufficient to precipitate more than the total amount of calcium, without the aid of the salts naturally in the blood. Mixtures of carbonate and phosphate present the usual char- acter, each salt being precipitated in its own peculiar form and being thus easily identified (Fig. 36). PRECIPITATION IN HYALINE CARTILAGE EXTRACT. An extract of fresh hyaline articular cartilage was prepared by mashing it up in distilled water, allowing the mixture to stand a short time, and then filtering it and using the clear solution. The same concentrations of salts were used here as in other cases, but the resulting precipitate was very much less in amount. Precipi- tation occurred only in the area of contact of the solutions, was in 292 JAMES CRAWFORD WATT. a finely divided state, showed some tendency to form crystals, but mostly was in form of granules and very small spherules. There were no large spherules whatever. The total amount of deposit would lead one to think that very weak solutions had been em- ployed instead of the usual ones. Hyaline cartilage extract thus seems to decrease the amount of precipitation and to retain what deposit is formed in a very finely divided state. Addition of the various accessory substances used in previous cases had no effect, even lecithin apparently being powerless to produce the large spherules. These facts may have some significance, for here we have an extract of a tissue which although apparently similar to other carti- lage which: does ossify, and although situated in continuity with ‘ossified tissue, yet retains a matrix free from a deposit of calcium salts. Its extract forms the most unfavorable medium yet found in which to produce precipitation of calcium salts. Solutions of equal concentration in other media give three to four times the amount of precipitate found here. The experimental facts, it seems, are in complete accord with the nature of the tissue, which is unfavorable to the deposition of calcium salts in it. INFLUENCE OF HypDROGEN-ION CONCENTRATION. Loeb’s method was employed of treating gelatin to obtain it with a definite hydrogen-ion concentration. To 0.25 gm. gelatin in a test tube were added 20 c.c. of cold acid solution of a definite strength, and this was allowed to stand for 30 minutes, during which time the gelatin swelled remarkably, but did not dissolve. It was then filtered and washed on the filter paper four times with 25 c.c. of cold distilled water at 5° C. to remove all free acid. The gelatin was then transferred to a test tube standing in hot water, where it melted, and the volume was made Up to, 25 (Cie: with hot distilled water, thus giving a one per cent. gelatin solution. Two tubes for each strength of acid were prepared. On cooling the pH of each pair was determined approximately by Clark’s colorimetric method, and then one tube of the pair was rendered a M/to calcium chloride solution, and the other one a //4 sodium carbonate solution, and these two solutions were placed as usual in a cell to react. Se CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 293 There was also an alkaline series made similarly to the above. In each series there were ten grades as follows: a a Acid series treated with ad ; } sets HCl. HOW 32) 164) 8192 Alkaline series treated with ie ; ie uf see ue NaOH 16 32) (64 8192 The pH of the acid series of washed gelatins in one per cent. solution ranged from 3.2 to 6.0. Untreated gelatin, both in sheets and in powder, which had been used for all the previous experi-' ments was found to possess a pH of 6.0. 7 The weakest alkaline member of the series registered 6.2 and the fifth member of the series a pH of 9.0, above which no record was kept. The amount of precipitate visible to the naked eye was less in the acid series than the alkaline one. It was least of all in those specimens where the pH was in the neighborhood of the isoelectric point of gelatin, which occurs in the acid series at 4.7, and which marks the point where the gelatin is most insoluble and most inert. In these slides there was practically no diffusion and spreading of the precipitate, only a narrow band of deposit occurring along the line of contact of the reacting solutions. Such a condition would lead one to infer that the salts in solution are intimately bound to - the gelatin. | It was also noted that crystals (Figs. 17, 18, 19) were abundant, and very often large, in the various acid gelatins. Crystals were also present in the weaker alkaline specimens, but in gelatin treated with M//2048 NaOH or stronger alkaline solutions (Fig. 23) not a single crystal has yet appeared in a period of six months since the preparation of these slides. Here the precipitate is still in the form of very small spherules, often so densely packed as to prevent Brownian movement, but elsewhere still actively in motion, which has gone on continuously for six months without a change, and - gives every indication of continuing indefinitely. The larger of these spherules, approaching one micron in diameter, have a slow movement and so can be well observed, and have given definite proof of what was already suspected from a study of other slides, _ 294 JAMES CRAWFORD WATT. - namely, that the so-called spherules are in reality flattened discs. During their vibratory movement many of them have been ob- served to slowly roll over and a narrow, thin edge is presented to the view instead of the circular outline. The appearance reminds one very strongly of a red blood corpuscle. The largest and clearest spherules (Figs. 19-22) were formed in gelatin whose pH was between 4.7 and 7.5— that is, all in gelatin on the alkaline side of the isoelectric point—but in specimens whose reaction was either weakly acid or weakly alkaline, the neutral point being 7.0. This is the zone in which is obtained the forma- tion of the largest spherules. Untreated gelatin, used for all the previous series of reactions without regard to hydrogen-ion con- centration, has a pH 6.0 and so is within this zone. The pH of human blood and many tissues also lies in the alkaline part of this zone. It is interesting to note that in strong acid members of the series precipitation is rapid, and while spherules may early be present, the deposit in the course of a short time becomes entirely crystalline in structure. In strongly alkaline members precipitation is very complete, but the deposit remains for an indefinitely long time in the form of small, discrete spherules, which do not grow or fuse, and which continue to exhibit Brownian movement. No crystals appear in these latter cases, even after the lapse of many months. In the intermediate zone of weak acid, neutral, or weak alkaline gelatin both spherules and crystals appear in quantity, and the spherules are very large. In the part of this zone in which the hydrogen-ion concentration of blood and many tissues lies the deposit is mostly in the form of small spherules and the tendency to crystallization is very poor. This is essentially a colloidal phe- nomenon and seems significant in view of the fact that it has fre- quently been found that certain drugs in colloidal form have a much better therapeutic effect, have a much less damaging effect on healthy tissues, and are much more pleasant in their administra- tion to the patient. As an example of this, compare the actions of silver nitrate and of argyrol, a colloidal silver vitellin. The colloidal preparation much more nearly approaches the condition of the tissues themselves and the reactions ought to be of a more col- CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 295 loidal character, which I have shown is the natural condition with the hydrogen-ion concentration on the basic side of the neutral point. To the acid side of this point reactions tend to the ordinary chemical crystalline variety. DISINTEGRATION OF SPHERULES. A most remarkable phenomenon was observed during the last two months in some of the slides, where spherules have been ob- served in process of disintegration and transformation into crystals. This was first noticed in the acid gelatin series in a slide where the gelatin had a pH of 4.9 and where for the whole of six months appearance had been constant. This case (Fig. 31) had offered one of the most perfect exam- ples of large spherules, which were of perfect outline, glassy trans- parency, absolutely clear, and showing no indication at all of their internal structure. These spherules were formed in half an hour and had attained their full growth in less than two days, and for six months presented the same appearance, with no indications of a change. It happened at the end of this period that they were not examined for about one month, when it was discovered that some of them (Figs. 19 and 33) had disappeared, others were dis- integrating, and all still present were opaque, while in this area which had not previously shown a single crystal there was now a number of large crystals. Other slides were examined for similar conditions without avail, but a month later several of them showed signs of a change, and by examination of these at intervals a complete series of stages in the transformation has been obtained and a complete description of it can now be given. The first sign of a change is in the occurrence of a very fine opaque central spot of a bluish tint (Figs. 20, 21, 24a, and 32) in the otherwise clear spherule. This area increases a little in extent, and the next stage then is ushered in by the appearance around this center (Fig. 24b) of a faint nebulous haze, which becomes more opaque and finally seems to consist of a fine radial striation (Fig. 24c) which extends farther and farther, until it reaches the pe- riphery of the spherule, which is now an entirely opaque white 296 JAMES CRAWFORD WATT. body. At this stage some spherules are uniformly opaque (Fig. 24d), others show a marked radial striation (Fig. 24e). The final step in the change now occurs with a very gradual. disappearance of the spherule, apparently by dissolving into the surrounding solution. It decreases in diameter and also in thick- ness until the remnant, diminished very much in size, appears like a nebulous ghost of the former spherule and eventually fades com- pletely from view. During the solution the opaque center of many spherules seems to dissolve early (Figs. 24f and 32), converting the body into a ring. | In those cases where the spherule is uniformly opaque (Figs. 24d and f) solution is very uniform at the surface, the contour of the spherule remaining perfect while the size contracts. Where there is radial striation, however (Fig. 24e and g), the periphery dissolves unevenly, giving the spherule a rough, rather ragged surface. Coincident with the beginning of the process of dissolving of the spherules, crystals occur in these areas which formerly were en- tirely composed of spherules, and which for several months’ dura- tion had not shown a single crystal. These crystals grow steadily both in size and in number (Figs. 19, 20, 21), while the spherules disappear. The passage of the material from the spherule into solution is evidently very transitory, it coming out again imme- diately to be deposited in the crystal. What is the interpretation of this phenomenon? It seems to me that there is a very evident one, and that this is the same process very much delayed in starting, and very much protracted in its action, which occurs rapidly in ordinary aqueous solutions. It will be remembered that in precipitating calcium carbonate in pure water it is at first in a granular, amorphous form, which later re- dissolves and comes out of solution a second time as spherules and crystals, which grow rapidly, the spherules all being converted into crystals during growth. _ Again in colloidal solutions the same process happens, but is - delayed somewhat, taking hours or days instead of minutes. In such cases an area of spherules seen one evening after a precipita- tion, when examined next morning is found to consist of crystals. CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 297 Apparently in some of these cases at least, as Brownian movement ceases and the particles grow in size, the spherule is transformed directly into the crystal. In the degeneration of the large spherule it seems to me we have a similar phenomenon to that of the other cases. The large spherules could correspond to the amorphous particles of the aqueous solution, fused into a large mass, the redissolving occurring very late to form the final crystal. There is evidently a very strong directive force toward the formation of crystals even in the col- loids, and though interfered with and hindered by them for a long period, it nevertheless conquers. What it is that initiates the change after the deposit has remained stationary for many months is problematical. There was no infection in the gelatin, no change in its appearance, or any other fact to give a clue. Some slides in other series in both gelatin and albumen also show evidence of some change in the spherules, which in view of the above facts may be expected to progress farther. As far as I know this remarkable transformation of the pre- cipitate has never been described before. The further investiga- tion of this phenomenon is outside of my province and belongs to _ the field of physics and chemistry, where it will provide some very interesting problems which I hope some competent investigator will undertake to solve. . | A second method of disintegration (Fig. 6) has also been ob- served in a lecithin solution, where some large spherules with very marked striation lost this striation after the lapse of several months and became dully opaque. Certain clear radii then appeared along which cleavage occurred, splitting the spherule into sectors which finally separated, giving triangular crystals. NuMBER oF Forms ASSUMED BY PRECIPITATE. It is to be remarked that in no slide was it possible, after Brown- ian movement ceased, to obtain only a single form of deposit. There were always two, and in many cases several forms occurring, and this applies equally to pure aqueous and to colloidal solutions. This multiplicity of form in the precipitate has also been noted by Biedermann and is due doubtless to differences in the relative con- 298 JAMES CRAWFORD WATT. centration of the reacting salts and in the diffusion currents in different parts of the solution. Side by side may be found an area of crystals and of perfect spherules, or both forms may be seen mixed in one area. In passing from the sodium carbonate end of a cell across the precipitate to the calcium chloride end there is often seen a marked change in shape of the deposit, forms present to one side of the line of contact of the two solutions not being present on the other. Spherules may occur all to the one side, but follow no rule as to which they select. Transition forms occur from one area to an- other. Orde found similar conditions on sectioning plugs of coag- ulated albumen with which he had sealed the ends of capillary tubes containing solutions which gave precipitates in the plugs when the sealed ends of the tubes were immersed in certain other solutions. The multitude of forms exhibited in the precipitate of calcium carbonate in the slides seemed remarkable, but on referring to Goldschmidt’s “ Atlas der Krystallformen,” I found illustrations for 2,544 modifications of the crystal form of calcite which have already been observed. Many of the changes are slight, but all are quite evidently different, and among them may be recognized all the types of crystal form I have obtained. Spherules and their variations were not shown. PART 2. EXAMINATION OF BONE. TECHNIQUE AND MATERIAL. The ordinary method of examining bone by means of very thin ground sections, or by sections cut from decalcified bone, are use- less to give the structure of the matrix. A method just published by Bast, however, for the study of bone cells im situ, without cut- ting sections, is equally applicable for the examination of the matrix and was here employed. Bast worked with the parietal and ethmoid bones of mice and young rats, rabbits, cats and dogs, the nasal conchz of these ani- mals, pieces of the ethmoid in man, and other bones thin enough to be transparent. These were mounted whole. The specimens were fixed in 95 per cent. alcohol, washed in water, stained for CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 299 8 hours or more in diluted aqueous solution of Gentian violet, dehydrated in alcohol, cleared in benzol, and mounted in balsam. To stain the bone cells only, the periosteum was left on the specimens, and this was later removed by dissection under a bi- nocular microscope while the bones were in benzol. In this way the matrix remained clear, while if the periosteum was removed previous to staining, the matrix stained also. This method was used by me, but as the matrix was the main _ object of study, most of my preparations were not stained, but were mounted for observation clear, in three different ways: first, without fixation, in a shallow cell filled with normal saline; second, in a shallow cell of glycerin; and third, fixed in 95 per cent. alcohol, then absolute alcohol, cleared in benzol, and mounted in balsam. A cover glass to which some of the precipitate of calcium car- bonate was adherent in the form of both spherules and crystals was passed through all the processes necessary to staining above and mounted in balsam. There was no change to be seen in the precipitate, so it can be inferred that calcium salts in the bony matrix, if in this form, would not be altered by any of the processes. The specimens of bone used were parietal and ethmoidal bones and nasal conche of the mouse, frontoparietal bones of the frog, and splinters and thin sections in various planes from the femur, tibia, humerus, radius, ulna and scapula, of frog, mouse, guinea pig, dog, and human fceetus. The frog and mouse were young, so that their bones were in process of growth, and if there is any visible difference between new formed and old matrix it should be evident here. Thin pieces of the long bones cut with a sharp knife to take in areas of the total thickness of the shaft, or to show transi- tion from the bony shaft to the cartilaginous epiphysis, were taken, splinters were clipped off long bones, small pieces were crushed, and all these were mounted as described above. The whole of a concha, scapula or parietal, if small, was mounted as one specimen. In all cases the periosteum was carefully peeled off and was mounted alongside of the bone so that examination of it for newly formed matrix could also be made. 300 JAMES CRAWFORD WATT. RESULTS OF EXAMINATION OF BONE. Both bright and dark field illumination were employed and specimens of bone, when thin, were quite transparent and easily examined. In bone that is well formed, whether from adult, or young individual, or foetus, the appearances are similar. The bone cells and all their processes in the canaliculi are easily followed and appear as Bast has described them. The matrix round about them is homogeneous, transparent, and almost glassy in appearance, and gives no indication (Fig. 37) of being formed of separate parti- cles. If it was originally formed in such a manner, there has been complete fusion, so that there is no evidence of a precipitation. In fact, the whole appearance bears out the statements of text-books of histology that, although two thirds of the bony: matrix is com- posed of inorganic salts, they are so intimately blended with the organic material that there is no visible evidence of their presence. In the case of growing long bones, such as those of the limbs, sections were taken across the shaft to include the newly formed growing periphery just under the periosteum, and also sections in the length of the bone to include the end of the diaphysis, the epiphyseal plate of cartilage, and the bony epiphysis. These sec- tions, however, were disappointing, as even at the transitional areas any tissue that could be identified as bony matrix appeared homo- geneous, so that no evidence of precipitation could be obtained here. In the case of the developing diaphysis of the long bones of the limbs in human foetuses, however, there were definite, discrete particles to be seen. In Quain’s Anatomy there is a fine de- scription by Schafer of the appearances of developing bone with which my findings entirely agree. Just ahead of the advancing bone of the diaphysis is an area of change, where the tissue is no longer cartilage, but is not yet true bone. Here we find definite fibres (Fig. 38) like those of white fibrous tissue, and known as Sharpey’s fibres, extending from the bony material forward into the cartilage, and in these fibres in the narrow area immediately in advance of the true bone are numerous very fine granules which appear similar to the material of the matrix in well-formed bone and so are evidently the bone salts. They are crowded in the fibres just ahead of the actual bone, and in the bone itself we find as CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 301 the homogeneous appearance due to the fusion of these discrete particles into one mass. Pacchioni has also described the first ap- pearance of lime salts in bone in the form of fine granules which later form a homogeneous mass. The possible origin of these particles either by precipitation or secretion will be discussed later. . PART 3. DISCUSSION OF RESULTS. : Concerning some of the physical phenomena of the precipitation reactions studied, remarks have been made at appropriate places in the previous pages describing the work and need not be repeated here. There are a few points to be constantly borne in mind in con- sidering the question of ossification, which have been very clearly demonstrated by Wells. The most important is the fact that there is a very definite and fixed ratio between the amount of calcium carbonate and calcium phosphate deposited in bone. The second point is that this same ratio is equally true in areas of calcification in tissues other than bone. The third fact is that the processes of ossification and of calcification are essentially the same, the kind of matrix or tissue in which the deposit occurs differentiating the two processes. THEORIES REGARDING DEPOSITION OF CALCIUM SALTS. The reactions with calcium phosphate show it to be invariably precipitated in a granular form so that if it were visible in bone it should be evident in a finely divided amorphous form. As this salt forms over eighty per cent. of the inorganic material, it ought to be readily demonstrated, but this is not the case, because the matrix is clear and homogeneous, not granular, except at the ad- vancing edge of ossification in developing foetal bones. It has been shown that in mixed solutions calcium carbonate is not prevented, by the presence of the phosphate, from coming out in typical spherules or crystals, so that it might be expected that the carbonate would show in one of these forms in bone, in which it forms fifteen per cent. of the salts. This expectation is further strengthened by the fact that typical spherules have been found in the shells of invertebrates, which are mostly calcium carbonate, by 302 JAMES CRAWFORD WATT. Rainey and by Biedermann. The dentine and enamel of the teeth, Rainey believed he had demonstrated to be composed of spherules, forming in rows and coalescing. Neither in old bone nor in new, nor in a transitional area transforming from cartilage to bone, have I seen either a single crystal or a single spherule. Previous work has demonstrated that spherules can always be found in calcium carbonate precipitated in colloids. This forma- tion is favored by the occurrence of the reaction gradually and slowly, and also by the presence of lecithin, which is found in both blood and bone, so that all the conditions in the matrix of the bone are favorable to the formation of spherules if precipitation occurs. The hydrogen-ion concentration is also favorable to formation of large spherules. With optimum conditions for their occurrence, and with a complete absence of these bodies, the interpretation 1S that the salts found in the bony matrix are not deposited in it by simple precipitation from a double decomposition of salts in the blood or the tissue. This is purely negative evidence, which, of course, is not of the same value as positive evidence, but it seems to me forms a very strong point against the simple precipitation theory. It must be noted, however, that if calcium carbonate is very small in amount.compared to the amount of phosphate precipitated that spherules may not be in evidence, but the whole precipitate will be granular, with coarser granules, resembling minute spher- ules and interpreted as carbonate interspersed among the masses of finer granules formed by the phosphate. It must be empha- sized that homogeneous masses such as seen in the bony matrix never occurred in any of my experiments, but precipitates, no mat- ter how long they were kept, remained composed of distinct dis- crete particles. Another point against this theory is that it supposes that a soluble salt of calcium circulates in the blood to be precipitated as phosphate and carbonate in the bone. With the large amount of sodium carbonate and sodium phosphate in the blood, calcium car- bonate and phosphate should be formed at once, here and not in the bone. But calcium is present in the blood in a greater amount than calcium carbonate and calcium phosphate are soluble in water. CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 303 And so it might appear that calcium can not be carried in the blood in the form in which it is later deposited in the bone. Pauli and Samec, however, have shown that difficultly soluble salts like cal- cium carbonate may have the amount of their solubility in water increased to over seven-fold in an albuminous solution, and so it is possible that the bone salts, formed by interaction of the blood salts and calcium absorbed into the blood as calcium chloride, might still be carried in solution in the blood and deposited in the bone. This-is closely related to the views of Barillé, whose theory is a step in advance of this, and is strongly supported by such work as that done by Wells. According to this view the calcium is in the blood in the form of a double salt formed here, tribasic calcium carbonophosphate, P,O,Ca,H, : 2CO,(CO;,H),Ca. This is soluble in the blood concentration of carbon dioxide. Infiltrating tissues like the bony matrix where the carbon dioxide content is lowered, this is precipitated and immediately breaks up into calcium car- bonate and dicalcium phosphate, which latter salt is converted into calcium phosphate, giving the final proportion found in bone of 15 parts calcium carbonate to 85 parts phosphate. Even by this method of precipitation, however, one might expect to be able to see the salts in the matrix, and not being able to do so argues in favor of their being combined with the matrix. The most impor- tant point brought out in this theory is a clear consistent explana- tion supported by experimental evidence, as to the fixed proportion in the amoumt of the two calcium salts, the carbonate and phos- phate, which is always found in their deposition. It is interesting to note here the following facts regarding carti- lage. Wells has shown that there is an almost specific power in cartilage for the absorption of calcium salts. My work is in agree- ment with this, for in cartilage extracts there was obtained the least amount of precipitation obtained in any of the experiments, thus showing the power of the substances in cartilage to hold on to the calcium. In spite of this affinity for calcium, however, microscopic evidence of developing ossification in bones preformed in cartilage shows us that the deposition of calcium is never directly in a cartilaginous matrix, but that the cartilage preceding the bone is destroyed and replaced by the organic bony matrix which is of 20 304 JAMES CRAWFORD WATT. a fibrous character, and this then has the calcium salts deposited in it. The final possible explanation of the deposit of bone salts in the matrix is the secretory one. Wells and others have been forced to the conclusion that there is a selective or specific action of tis- sues, such as cartilage or membrane about to ossify, by which calcium salts are absorbed or otherwise taken into them, while apparently entirely similar areas of cartilage and membrane do not ossify at that time or at any later time. From this view of selective action it is not a far step to that of secretory action, explaining the ossifying process as a secretion of the calcium salts by the bone forming cells. It is already generally admitted that the matrix is a product of the bone cells, and if this is true, it is quite logical to assume that the calcium salts are taken by the bone cells from the blood and passed on into the matrix. The difficulty in this theory has been to account for the amount of phosphate in the tissues. The calcium was looked upon as being derived from that carried in the blood, but the phosphate was explained as being due to cellular activities, derived either from the nucleus or from cellular degeneration. Another difficulty here, again, is to explain the fixed ratio of carbonate and phosphate. This, of course, could be looked upon as due to the carbon dioxide content of the tissue, or as due to the activity of the cell, a definite proportion being found in the secretory products here just as in other organs of the body. It seems to me, however, that the nearest approach to the truth is to be obtained by combining two views. The bone salts, I be- lieve, reach the tissue as the soluble double salt, tribasic calcium carbonophosphate, which by its constitution provides for the tissue both calcium carbonate and calcium phosphate, each salt in the proper proportion found in bone. I do not believe, however, that deposit here is by precipitation, for in the bone, owing to the col- loidal matrix, the hydrogen-ion concentration, and the presence of lecithin, all conditions are most favorable to a visible and char- acteristic precipitate, showing granules and spherules, but there is no visible evidence whatever of these bodies. The only evidence in favor of precipitation consists of the following two sets of facts: CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 305 First, Schafer and Pacchioni have described the first appearance of bone salts in developing bone in the form of minute granules which quickly coalesce. This I have confirmed. This evidence could also be interpreted as showing the presence of the salts to be due to secretory activity. Secondly, Olivier in a tumor of the breast and Pettit in a tumor of the maxilla and also in a renal cyst saw and made illustrations of calcospherites. These bodies can be explained only as a precipitation phenomena, but they do not afford sufficient evidence to prove that this is the process whereby the bone salts are deposited. And so these salts, after reaching the bone in correct proportions as the double salt above mentioned, must be deposited in the matrix by some more subtle process than ordinary precipitation. Here we have recourse to the secretion theory. The bone salts in their proper proportions having been brought to the cell by the blood, are taken by the cell and secreted as an integral part of the matrix in combination with the ossein, the organic base, which it is generally admitted already, isa product of the cells. This will produce a matrix uniform in appearance, with salts invisible, and provides the proper proportion of each salt, as the cell is in its turn provided with the two salts in that exact amount. Histological evidence shows that the fibrous matrix appears first in the newly forming bone and impregnation with the bone salts begins immedi- . ately, the salts showing as fine granules at first, and coalescing as their quantity increases, to form a homogeneous matrix. REVERSIBILITY OF CaLcrum REACTION. There seems to be evidence that this process of secretion is reversible, the bone cells being able to take up the calcium salts again. Wells and others have shown that calcium is removed from the bones in cases where there is a great demand for it in the body, as in pregnancy, or where it is being steadily lost by the body, for instance by way of a pancreatic fistula. Also, the osteoclasts re- sponsible for the normal resorption of bone are supposed by some authors to be only osteoblasts which have changed their function. And further, where a graft of either living or dead bone, or a bone plate and bone screws are used to repair an injury, a process known 306 JAMES CRAWFORD WATT. as creeping replacement occurs whereby this material is gradually absorbed and replaced by new bone. It has been shown by Gallie and Robertson that it is the osteoblasts which invade the graft, gradually absorb the old matrix and bone salts immediately sur- rounding them, and then replace this by newly formed substance. If the bone salts were laid down by precipitation of tribasic calcium carbonophosphate due to a change of the carbon dioxide content, it seems to me it would be impossible for this precipitate to be removed again as easily as Wells states it is, in response to the body’s need for calcium. Would pregnancy or a pancreatic fistula so change the carbon dioxide content of the bone, as to make possible a reversal of the conditions that brought about the pre- cipitation, and so either provide for increased solubility of the two salts or else make possible the recombination of the carbonate and phosphate of calcium into the double salt, with its consequent solu- tion and carrying off in the blood? It seems to me that the carbon dioxide content in the bone will not vary enough to permit of such a reversal, as to raise it would indicate a high activity in a tissue whose metabolism is low. This low-grade activity is given as a reason for obtaining the precipitate. The taking of the calcium salts out of the bone probably depends on some balance that has to be maintained between the blood, the osteoblast, and the bony matrix. If the blood is rich in calcium, the cell can take it and pass it on into the matrix. If the blood is poor in calcium, the cell to maintain its relation in balance to the blood takes from the matrix to add to the blood. This is an instance comparable to the relations between the blood and its dextrose, and the liver cell and its stored glycogen, to maintain a certain carbohydrate balance. This view also follows logically upon that of Wells, who regards the bony matrix as a great reserve store of alkaline bases for the body, where the calcium salts are in a state of flux, being constantly added to or drawn upon according to the needs of the body. Changes in the bones are much greater in extent and more rapid in their occurrence than is generally believed. It seems to me that this idea of the reversibility of the direction in which calcium salts CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 307 pass through the tissues is a most valuable one for the proper com- prehension of some of the processes of growth, repair of fractures, and changes going on in rarefaction of bony tissues. SUMMARY. 1. Calcium phosphate precipitated in water or in colloidal solu- tions is constantly granular and amorphous in character, and apparently uninfluenced by the nature of the solution. 2. Calcium carbonate precipitated in water shows a great di- versity of crystalline form; in colloidal solutions it shows two main forms, an irregular crystalline one and a spherical. 3. Mixtures by simultaneous precipitation of both salts in the same solution show each salt separating out independently and the part of the deposit formed by each can be easily identified. 4. Spherule formation by calcium carbonate is a typical reaction in colloidal solutions. 5. Spherules and crystals are influenced in shape, size, number, and internal structure by a variety of substances found normally or pathologically in the blood, most notably by lecithin which favors the formation of large spherules, and by acetone which increases the rapidity and extent of precipitation. 6. The character of the deposit of calcium carbonate is influenced by the hydrogen-ion concentration of the colloidal solution, being most crystalline in acid media, all in the form of spherules in strongly alkaline media, and mixed in form in solutions neutral or nearly neutral. : 7. Large spherules, after persisting for months may undergo a sort of degenerative process, with change of internal structure, after which they dissolve and the material forming them is laid down in a crystalline form. As far as I know this phenomenon has never been previously described. 8. In fresh bone of various animals examined in various ways there is no visible microscopic evidence of the bone salts, although they form two thirds of the mass of the matrix. The inference is that the bone salts are not deposited in the matrix by simple precipitation, for the conditions are such that, if precipitated, granules, spherules, and crystals should be visible. 308 JAMES CRAWFORD WATT. g. In the rapidly developing fcetal skeleton the first appearance of bone in the matrix is in the form of fine granules or globules which quickly fuse or coalesce to form a homogeneous mass. This might be interpreted according to the bias of the observer as sup- porting evidence either of precipitation or of secretion of the salts into the matrix. f 10. The view advanced by Barillé and supported by Wells’s work, that calcium is carried in the blood as tribasic calcium car- bonophosphate, is probably correct, as it furnishes the bone salts in the proper proportion, but their view of its deposit in the matrix as a precipitate due to change in concentration of carbon dioxide does not appear correct in view of the fact that no precipitate of the bone salts is visible. 11. The theory that the salts furnished by the blood are taken by the bone cells and secreted by them along with the matrix seems reasonable in view of the condition found in the matrix. 12. The action of the osteoblasts seems to be reversible, they being able to absorb or take up the calcium salts again out of the matrix. Bik. In concluding this paper I wish to acknowledge my indebtedness to others by thanking Professor J. Playfair McMurrich for special apparatus and for valuable references, also the following for pro- viding material for investigation: Professor W. H. Piersol, Dr. R. D. Defries, Dr. Banting, Dr. Simpson, Dr. Dafoe, and Dr. Low. BIBLIOGRAPHY. Barillé, A. ’04 ~De l’action de l’acide carbonique sous pression sur les phosphates métal- liques; combinaison (carbonophosphates) ou dissolution, applications diverses. Journ. de Pharm. et de Chimie, Ser. 6, T. 19, p. 14, 71, 196, 245, 295. "10 Role des constituants de la dissociation de la carbonophosphate trical- cique dans la genése du tissu osseux et des diverses concrétions a base de phosphate et de carbonate de calcium. Journ. de Pharm. et det@himiey Sere. ieee ans: Bast, T. H. ’ 21 Studies on the Structure and Multiplication of Bone Cells facilitated by a new Technique. Amer, Jour. of Anat., Vol. 29, pp. 139-157. ‘eh SIS CALCIUM PHOSPHATE AND CALCIUM CARBONATE. 309 Biedermann, W. ’oz Ueber die Bedeutung von Krystallization prozessen bei Bildung der Skelette wirbelosser Tiere, namentlisch der Molluskenschalen. Zeit- schr. fiir allgemeine Physiol., Bd. I., S. 154-208. 02 Untersuchungen tber Bau und Entstehung der Molluskenschalen. Je- naische Zeitsch. fiir Naturwiss., Bd. XXXVI., S. 1-164. Bowman J. H. ’06 «6A Study in Crystallization. Journ. of Soc. of Chem. Indust., Vol. 2s, Pp. 143. Clark, W. M. ’21 The Determination of Hydrogen Ions. Williams and Wilkins Co., Baltimore. Gallie, W. E., and Robertson, D. E. 18 The Transplantation of Bone. Journ. of Amer. Med. Ass., Vol. 70, pp. 1134-1140. "19 The Repair of Bone. Brit. Journ. of Surg., Vol. VII., No. 26, pp. 211-216. , Goldschmidt, V. : 13 Atlas der Krystallformen. Vol. 2 and Atlas Vol. 2. Winters, Heidel- berg. Gradwohl, R. B. H., and Blaivas A. J. ’20 The Newer Weiibds ‘of Blood and Urine Chemietrys 2d Edition. C. V. Mosby Co., St. Louis, U.S. A. Harting, — "72 On the Artificial Production of some of the Principal Organic Cal- careous Formations. Quart. Journ. Micros. Science, New Series, Vol. XII., pp. 118-123. Holmes, H. N. 17 The Formation of Crystals in Gels. Journ. of Physical Chem., Vol. XXI., pp. 709-733. Lillie F. R. ’08 The Development of the Chick. Henry Holt and Co., New York. Link, H. F. . ’39 Ueber die erste Entstehung der Krystalle. Annalen der Physik und Chemie. Poggendorf, Bd. 46, S. 258. Loeb, Jacques. 18 Amphoteric Colloids, I-V. Journ. of Gen. Physiol., Vol. I. I. Chem- ical Influence of the Hydrogen Ion Concentration, p. 39. II. Volu- ? metric Analysis of Ion Protein Compounds; the Significance of the Isoelectric Point for the Purification of Amphoteric Colloids, p. 237. III. Chemical Basis of Influence of Acid upon the Physical Proper- ties of Gelatin, p. 363. IV. The Influence of the Valency of Cations upon the Physical, Properties of Gelatin, p. 483. V. The Influence of the Anions upon the Physical Properties of Gelatin, p. 559. Loeb, Jacques. %’290 Ion Series and the Physical Properties of Proteins. Journ. of Gen. Physiol., Vol. 3, p. 85. ’290 ©The Proteins and Colloid Chemistry. Science, New Series, Vol. LII., No. 1350, PP. 449-456. 310 JAMES CRAWFORD WATT. Olivier, E. ’95 Cancer gelatineux du Sein avec Corps Calcaires. Beitr. zur Path. Anat. und zur Allgem. Pathol., Bd. XVII. Ord, W. M. ’72, On Molecular Coalescence and on the Influence of Colloids upon the Forms of Inorganic Matter. Quart. Journ. Micros. Science, Vol. XII., pp. 219-239. Pacchioni D. 792 Untersuchungen uber die normale Ossification des Knorpels. Jahrb. f. Kinderheil. Berl., Bd. LVI., S. 327-340. Pauli, W., und Samec, M. 709 ~Ueber Loslichkeitsbeeinflussung von Electrolyten durch Eweisskorper. Biochem. Zeitschr., Bd. 17, S. 235-256. Pettit, A. ’97_ ‘Sur le Réle des Calcospherites dans la Calcification a l’Etat Pathologique. Arch. d’Anat. Micros., T. 1, p. 107. Rainey, George. ’57_ On the Formation of the Skeletons of Animals, and other Hard Struc- tures formed in Connection with Living Tissues. Brit. and Foreign Med. Chir, Review, Vol. XL., p. 343. : ’58 Precise Directions for making of Artificial Calculi, with some Obser- vations on Molecular Coalescence. Quart. Journ. Micros. Sci., Vol. VI., pp. 41-50. ’59 On the Structure and Mode of Formation of the Dental Tissues, ac- cording to the Principle of Molecular Coalescence. Quart. Journ. Micros. Sci., Vol. VII., pp. 212-225. 761 Some Further Experiments and Observations on the Mode of Forma- tion and Coalescence of Carbonate of Lime Globules, and the Develop- ment of Shell Tissues. Quart. Journ. Micros. Sci., New Series, Vol. I, Dp. 23-32. : Richards T. W., and Archibald, E. H. 7or A Study of Growing Crystals by Instantaneous Photomicrography. Amer. Chem. Journ., Vol. XXVI. Schafer, E. A. %12 Text-book of Microscopic Anatomy. Quain’s Elements of Anatomy, Eleventh Edition, Vol. II., Pt. I. Longmans, Green and Co., London _ and New York. Wells, H. G. 706 Pathological Calcification. Journ. Med. Research, Vol. XIV. (N. S., Vol. 9), p. 491. "10 «©6Calcification and Ossification. Harvey Lectures, 1910-11. Iz Calcification and Ossification. Arch, of Int. Med., Vol. VII., p. 721. 718 Chemical Pathology. Third Edition. W. B. Saunders and Co., Phila- delphia and London, CALCIUM PHOSPHATE AND CALCIUM: CARBONATE. 311 EXPLANATION OF PLATES. PLATE I. All figures are from camera lucida tracings and are of equal magnification, X 200, so that the relative size of precipitated particles in each case can be compared. Only forms which were common, and appeared in quantity have been shown. All illustrations are to show forms assumed by calcium carbonate precipi- tated from various solutions by the reagents listed below. Fic. 1. Crystals of calcium carbonate precipitated by sodium carbonate in distilled water. Fic. 2. Crystals of calcium carbonate precipitated by ammonium carbonate in distilled water. Fic. 3. Crystals of calcium carbonate precipitated by potassium carbonate in distilled water. Fic. 4. Deposit of calcium carbonate precipitated by sodium carbonate in distilled water containing 0.1 per cent. lecithin. Note small spherules, and small crystals, a tendency towards clumping, a very large spherule and a large rosette. ; Fic. 5. Deposit of calcium carbonate precipitated by ammonium carbonate in distilled water containing 0.1 per cent. lecithin. Spherules of various sizes, a marked tendency to fusion, and formation of large irregular masses are to be noted. Fic. 6. Deposit of calcium carbonate precipitated by potassium carbonate in distilled water containing 0.1 per cent. lecithin. Note a tendency to clump- ing, various sizes of spherules with some double forms. The last three forms indicate one process of disintegration of a spherule, by the occurrence of radial lines along which it splits into sectors, which thus become triangular crystals. Fic. 7. Deposit of calcium carbonate precipitated by sodium carbonate in a solution of egg albumen. Practically the whole precipitate was in form of spherules. ; Fic. 8. Deposit of calcium carbonate precipitated by ammonium carbonate in a solution of egg albumen. Spherules of varying size are seen, also some crys- tals. Fic. 9. Deposit of calcium carbonate precipitated by potassium carbonate in a solution of egg albumen. Many spherules are seen, and a tendency to- wards fusion is noted. Fic. 10. Deposit of calcium carbonate precipitated by sodium carbonate in a solution of egg albumen containing o.1 per cent. lecithin. Note the markedly increased size of spherules and crystals and the tendency to fusion. Fic. 11. Deposit of calcium carbonate precipitated by ammonium carbonate in a solution of egg albumen containing o.1 per cent. lecithin. Note the in- creased size of particles, and their tendency to fuse. Fic. 12. Deposit of calcium carbonate precipitated by potassium carbonate in a solution of egg albumen containing 1 per cent. of acetone. Crystals and spherules both occur. One crystal is seen fused with a spherule. Fic. 13. Deposit of calcium carbonate by interaction of calcium chloride in a 1 per cent, gelatin solution with sodium carbonate in a 4 per cent. gelatin. 312 JAMES CRAWFORD WATT. Fields are shown by various sized spherules, of crystals, and of spherules and crystals mixed. Fic. 14. Deposit of calcium carbonate by interaction of calcium chloride in 1 per cent. gelatin solution with ammonium carbonate in 4 per cent. gelatin solution. Spherules of various sizes are’ seen. Fig. 15. Deposit of calcium carbonate by interaction of calcium chloride in 1t per cent. gelatin solution containing 0.1 per cent. lecithin, with am- monium carbonate in 4 per cent. gelatin solution containing 0.1 per cent. lecithin. Note the increase in size of the spherules over those shown in Fig. 14. Three different degrees of fusion are illustrated in the double forms. Fic. 16. Deposit of calcium carbonate precipitated by sodium carbonate in a I per cent. solution of untreated gelatin. Fic. 17. Deposit of calcium carbonate precipitated by sodium carbonate in a i per cent, solution of gelatin previously treated with 14g HCl solution. Large crystalline precipitate. Fic. 18. Deposit of calcium carbonate precipitated by sodium carbonate in a I per cent. solution of gelatin previously treated with 145 HCl solution. Large, markedly crystalline deposit. IRIKER 5x0): Deposit of calcium carbonate precipitated by sodium carbonate in a zr per cent. solution of gelatin previously treated with 14,5 HCI solution. The field shown in this figure was composed for some months altogether of clear spherules and a few dumb-bells. The spherules here illustrated are now undergoing disintegration and in their place are appearing large crystals. Some debris is seen in one corner. One spherule is seen partly from the side, showing its true shape to be a disc. 1G. 20. Deposit of calcium carbonate precipitated by sodium carbonate in a I per cent. solution of gelatin previously treated with 14943 HCl solution. Disintegrating spherules and new crystals are seen. Fic. 21. Deposit of calcium carbonate precipitated by sodium carbonate in a i per cent. solution of gelatin previously treated with M/8192 NaOH solu- tion. Newly formed crystals and various stages of disintegrating spherules are shown. Fic. 22. Deposit of calcium carbonate precipitated by sodium carbonate in a 1 per cent. solution of gelatin previously treated with M/4006 NaOH solution. ENG, 23}. @Peposit of calcium carbonate precipitated by sodium carbonate in a 1 per cent. solution of gelatin previously treated with M/2048 NaOH solu- tion. Precipitate is all in the form of small spherules which have been in constant Brownian movement for six months. Fic. 24. A series of spherules showing various stages in the disintegrative changes which occur before they redissolve. a, Spherule with small central opaque area. b, Spherule with opaque radial striation extending from central spot. c, Spherule with further extension of opaque area. d, Spherule almost completely opaque. Narrow, bright, clear periphery. e, Spherule striated right out to periphery. f, Spherule like that in d, dissolving both at periphery and in center in a regular manner. g, Spherule like that in e, dissolving at an irregular rate at the periphery. h, 7, Remains of dissolving spherules, and debris of some that have fallen apart. w * PLATE te ES = x a Ce) > z r Ww - a =) o a < 2 o ° a 9 a 314 JAMES CRAWFORD WATT. JPALYNitio) 2) Photomicrographs all taken at a standard magnification, X 200, with dark field condenser in use. Fie. 25. Crystals of calcium carbonate precipitated by ammonium carbonate in water. Fic. 26. Calcium carbonate precipitated by sodium carbonate in a solution of egg albumen containing N/1o sodium chloride. Fic. 27. Calcium carbonate precipitated by potassium carbonate in water containing 0.1 per cent. lecithin. Note one double spherule and three large single spherules of which the two larger show indications of the radial divi- sions which will result in splitting the spherule into crystals of triangular out- line. Many small crystals and spherules seen. Fic. 28. Calcium carbonate precipitated by ammonium carbonate in water containing o.1 per cent. lecithin. Fic. 29. Calcium carbonate precipitated by potassium carbonate in a solu- tion of egg albumen containing 1 per cent. ethyl alcohol. Fic. 30. Calcium carbonate precipitated by potassium carbonate in a solu- tion of egg albumen containing 0.05 per cent. urea. Fic. 31. Calcium carbonate precipitated by sodium carbonate in a I per cent. solution of gelatin previously treated with 1/1024 HCl solution. Clear spherules, some seen from side appearing disc-like. Fic. 32. Calcium carbonate precipitated by sodium carbonate in a 1 per cent. solution of gelatin previously treated with 1/2048 HCl solution. Some crystals are present. Spherules begin to show an opaque central dot, the first indication of a coming disintegration. Fic. 33. Calcium carbonate precipitated by sodium carbonate in a 1 per cent. solution of gelatin previously treated with 1/512 HCi solution. This field was formerly all spherules, now some crystals are seen. Spherules are far advanced in disintegrative process, are completely opaque, and are dissolvy- ing, some appearing only as ghosts of their former selves. Fic. 34. Calcium carbonate precipitated by ammonium carbonate in 1 per cent. gelatin solution containing 0.05 per cent. urea. Rings and striations seen in spherules. Fic. 35. Calcium carbonate precipitated by sodium carbonate in a I per cent. gelatin solution containing 0.05 per cent. urea. Field of very small mixed crystals and spherules. Fic. 36. Combined precipitate of calcium phosphate and calcium carbonate by action of sodium phosphate and sodium carbonate in dog’s blood serum. The calcium carbonate shows as seven spherules. The calcium phosphate shows as a cloud of very fine amorphous granules, which is the form in which it precipitated in all experiments. "8 > a x a ° > z F Wi a a =) a a < = fo} ° a ie) a E rr a s ‘ i . ’ f P ' ‘ i oy = , . ‘ 316 JAMES CRAWFORD WATT. IPPADE 3s Fic. 37. Photomicrograph of a small piece of bone from the growing fe- mur of a mouse. The bone cells show as small, oval, opaque bodies in the specimen. No discrete particles can be seen in the matrix. Magnification, xX 200, Fic. 38. Outline sketch made with camera lucida, of the growing bony end of the diaphysis of the humerus of a six months human foetus. Four zones can be distinguished. A, fully formed bone spicules (shown as solid black, marrow spaces white, bone and marrow cells not indicated) with calcium salts fused.° B, osteogenic fibres, impregnated with calcium salts in form of minute granules. C, newly forming osteogenic fibres, not yet impregnated with bone salts, and cartilage showing changes preparatory to bone formation. D, un- altered cartilage. Magnification, X 300, epee, a PLATE Ill. BIOLOGICAL BULLETIN, VOL. XLIv. O 9 2 9990090 LAS —— 0 999000 3 GG9 eake @ vg ¢ 99 J. GC. WATT. SMITHSONIAN INSTITUTION LIBRARIES OCU 3 9088 01228 1267