to genera found nowhere else, Gulick established the fact that in each mountain valley of the forest region of Oahii there is a great nuniher of local species (200. represented by 700 or 800 varieties) belonging to several genera, and that each of the twenty valleys contains one or more local varieties or species restricted to that valley. On tracing this wonderfully dilVercntiatcd assem- bly from valley to valley, it became apparent that a slight variation in the occupants of a valley as compared with those of the adjacent valley lie- comes more prunciunocd in the next or third valley, still more in the fourth, and so on. Thus he was able roughly to estimate the amount of di- vergence between the occupants of any two given valleys by measuring the number of miles be- tween them (Romanes). Gulick thinks the evo- lution of these different forms cannot be attribit- ed to dilTcrences in their external conditions. The forest area, covering one of the nunuitain ranges. in which the snails live, is about 40 miles long and 5 or miles wide. He states that the rainfall on the northeast side of the mountain is somewhat heavier than on the opposite side, and the higher ridges of the mountains are cooler than the valleys; but the valleys on one side of the range have a climate the same in every respect. The vegetation in the valleys differs somewhat from that on the ridges; but the vegetation of the dilTcrent valleys is much the same; the birds, in- sects, and larger animals are the same. Though, as .far as we can observe, the conditions are the game in the vallej's on one side of the range, each has a molluscan fauna differing in some degree from that of anj' other. He also adds that a genus is rejiresented in several successive valleys by allied species, sometimes feeding on the same, sometimes on different plant.s. In every such ease it appeared that the valleys that arc nearest to each other furnish the most nearly allied forms, and a full set of the varieties of each species pre- sents a minute gradation of forms between the more divergent types found in the more widely separated localities. After giving reasons tor the belief that this variation is not due to dif- ferences in their external conditions. Gulick con- cludes tlint the difference is due to a corre- .sponding difference in the time of separation of each variety, and also to what he terms 'cumula- tive segregation.' 'segregate lireeding.' and 'inde- pendent generation.' In his interesting essay on I'hysiolofiical .^election (1880 and 1807),' Ro- manes states that the essence of the principle consists in all eases of the diversifying effect of cross-infertility, whensoever and howsoever it may happen in partieilar eases to have been caused. (See Phy.sioi.ogicwl Selection. ) It is to be observed (hat this is but little different from Gviliek's 'cunmlative' or 'intensive' segre- gation. See Evolution, paragraph Factors of Evolution; Classific.mion op Animals, para- graph f^pecics.
After all the careful work done by Gulick. it is yet to be doubted whether the chief or initial factors in the wonderful specialization which has taken place in (he land shells of Oahu are not the result of migration into new regions, vary- ing in natural conditions. For the present, then, Wagner's factors of migration into areas with differincr conditions of life, and isolation, and the consequent prevention of intercrossing with the original or parent forms, may be accepted as the essential causes of the origin of perhaps two- thirds to one-half of existing as well as extinct species.
HiiiLiOGRAPHY. Ucbcr die Daruinisehc Theorie i» Bezug auf die ijeographische Verbreitung der Organimnen (Munich, 1808); The Darwinian Theory and the Laic of the Migration of Organ- isms (Leipzig, 1808; trans, by Laird, London, 1873); Leber den Einfliiss der gcugraphisrhen [sulirung und Colonirnhihinng auf die morpho- luiiisvhen Verundrrunyt ]i drr Orgiiiiisnicn (iMu- ni'ch, 1870); (iulick,' "On Diversity of Evolu- tion Under One Set of External Coiulitions," in loiiriial of the Linnean Socieli/ (London. 187'2); Divergent Evolution Through ('uniulatirc tlegre- gation (ib., 1888); "Divergent Evolution and the Darwinian Theory," in American .lourmil of Science (Xew Haven, .lanuary. 1890); "Inten- sive Segregation, or Divergence Through Inde- pendent Transformation," in Journal of the lAnncan Hocictji (London. 1800); Romanes, "Physiological Selection." in Journal of the Linnean Hocieli) (ib., 1880).
ISOLD, ISOLDE, ISOND, ISOUD. See ISEII.T.
ISOMERISM, i-s5m'er-iz'm. See Chemtstby;
Valency; Carhon CoMi-oiNn.s; Stebeo-Ciiem-
isTRY; Cyanic Acid; Ali.otropy.
ISOMORPHISM, rsi'.-mor'fiz'm (from iso-
miirphijuK. from (ik. imif, I'.so.s. equal 4- Ifii'f'fl,
morphc, fcum ) . The relationship existing be-
tween solid substances which arc similar in their
chemical composition and constitution, have simi-
lar crystalline forms, and are eapa!)le of forming
homogeneous mixed crystals ('solid solutions'),
and each of which is capable of growing in a
saturated solution of another, the latter then
forming a mantle around the first as a nucleus.
Ostwald proposes to define the relationship >c-
tween two isomorphous substances as the capac-
ity of one to cause immediate crystallization in
a supersaturated solution of another, just as a
crystal of any sibstanee is capable of causing
its ow^l sujiersaturated solution to crystallize
immediately. It nuist, however, be observed that
while the relationship of isomorphism certainly
exists, our knowledge of its intimate nature is
very vague, and neither of the above definitions
gives adequate expression to all of (he known
facts. The ordinary carbonates of calcium (cal-
eite), magnesium (magnesite). iron (siderite),
manganese (rhodoehrosite) , and zinc (smith-
sonite) all form crystals of the same crystal
system, and fir the most jiart of the same crystal
class (see Crystallocrapiiy-); and, further,
their corresponding interfacial angles approach
to the same values. Not all substances clo.sely
related in their chemical composition are iso-
morphous. and. exceptionally, substances which
h.ave no chemical relationship have similar sym-
metry of crystals and angles in close corre-
spondence. (Such an accidental resemblance of
the crystal forms of chemically unrelated sub-
stances is described as 'isogonism.') In the ca.se
of salts, the chemical component which seems
mainly to condition the synimctrv of the mole-
cule is the acid radical, the metal having less
influence upon the crystal's symmetry, though
affecting the size of crystal angles. (See MoB-
piioTROPiSM.) The met^als of a series of iso-
morphous salts are said to be isomorphous ele-
ments, and it is found that for the most part
