THE EVOLUTION OF SPINY LOBSTERS (PALINURIDAE): A STUDY OF EVOLUTION IN THE MARINE ENVIRONMENT

R. W. GEORGE Western Australian Museum, Beaufort Street, Perth, Western Australia

A. R. MAIN Zoology Department, University of Western Australia, Nedlands, Western Australia

Received October 18, 1966

The purpose of this paper is to evaluate to assess "significant" characters upon relationships in the light of morphological which to base our phylogenetic arrange­ and paleontological evidence, and to pro­ ment, we first tabulated the major external pose an evolutionary scheme which accounts morphological features which varied most for the present distribution of extant between genera. For some features no palinurid species. The spiny lobsters, be­ meaningful arrangement could be obtained. cause of their large size, commercial im­ For instance, the form of the abdominal portance, benthic habits as adults and sculpturing could be arranged as follows: pelagic habit as larvae, and their wide no sculpture, one transverse groove inter­ geographic and vertical distribution, are rupted medianly, one continuous transverse well suited for a study of evolution in the groove, four transverse grooves, seven marine environment. In 1946 Holthuis transverse grooves. However, other ab­ provided an excellent basic taxonomic re­ dominal sculpture patterns are found which vision of the Palinuridae and, sometimes are not easily fitted to this series, e.g. with colleagues, he followed this up with squamiform sculpture or longitudinal carina. more detailed studies of smaller groups Because of the inconsistency in the trend within the family, particularly in the genus of this feature, and because similar forms Panulirus. But some of the less well known of abdominal sculpture occur in many genera inhabiting deeper waters still require other crustacean groups which are unrelated revision and for this reason it has been to palinurids, and also because the partial necessary to include some taxonomic discus­ trend given above is not supported by the sion in the paper; this has been kept to trends of any of the other features, ab­ a minimum and will be reported on more dominal sculpture is considered to be a fully in an appropriate publication at a random development rather than a phy­ later date. The deficiencies in our knowl­ logenetic indicator. Two other features edge of palinurid systematics and evolu­ were also placed in this category (see p. tion have been realized by Glaessner (1960) 807). who stated (p. 42), "There is still some We were searching for morphological doubt about homologies and evolution in the characters which: (a) could be arrayed in Scyllaridea [Palinuridae and Scyllaridae] order and accommodate all the genera of ... for which groups more morphological the family in each arrangement; (b) were information on recent and fossil forms is consistent with one another in their trends; required." We have attempted to overcome (c) were based on features characteristic these deficiencies and to present speciation of the family and rarely, if at all, found mechanisms applicable to the marine en­ in other crustacean groups; (d) were vironment. judged most likely to have a selective advantage in certain environments. PHYLOGENETIC RELATIONSHIPS Four features examined fulfilled these We have based our phylogenetic scheme requirements but before presenting these entirely on external morphology. In order data (see p. 805) it is essential to present

EVOLUTION 21: 803-820. December, 1967 803 804 R. W. GEORGE AND A. R. MAIN

TABLE 1. Morphological relationships of living palinurid genera. The parallel trend of the Silentes and Stridentes for the same morphological characters is noteworthy. For the Stridentes genera the shape and armature of the antennular plate has also been included.

Distal joints of endopod of female SILENTES STRIDENTES Supra­ Eye second pleopod orbital posi­ Antennular Carapace process tion Penultimate Ultimate plate Cylindrical strong elevated full absent PANULIRUS broad, 2-4 vertical expansion large spines Cylindrical strong elevated full vestigial JASUS JUSTITIA triangular, almost expansion PALINURUS unarmed vertical Subcylindrical moderate horizontal moderate reduced PROJASUS PALINUSTUS triangular, slightly expansion unarmed elevated Subcylindrical small horizontal slight large LlNUPARUS narrow, flattened expansion PUERULUS unarmed a brief taxonomic statement on the genera shows that the two lines were present at of the family and to point out that the an early stage of palinurid development. presence or absence of a stridulating ap­ These structures have been described and paratus is regarded by us as indicating a illustrated by George and Grindley (1964) major divergence within the family. The who support the long-neglected proposal stridulating apparatus is formed by a of Parker (1884) to use "Silentes" for the lateral and anterior expansion of the anten­ non-stridulating group and "Stridentes" for nular segment, a dislocation of the interior the stridulating group. margin of the first free antennal joint from this segment, and a latero-internal expan­ PALINURID GENERA sion of the first free antennal segment. Holthuis (1946) listed eight genera in This expanded stridulating pad of the the family Palinuridae and since that time antenna is ribbed ventrally and when the one additional genus, Projasus George and antenna is moved, it rubs across a polished Grindley 1964, has been described. How­ ridge on the lateral margin of the anten­ ever we are of the opinion that Palinurellus, nular segment. Those genera which do one of the genera mentioned by Holthuis, not have the ability to stridulate have a should be removed from the palinurid list narrow, vertical antennular plate, the since it has obvious affinities with the antero-external margin of which serves as Scyllaridae and Polychelidae as well as the fixed articulation point of the antenna. with the Palinuridae. These affinities have The possession of a stridulating ap­ been discussed at length by de Man (1881, paratus is well known to be an advantage 1882), Bate (1881), and Boas (1882). to the species for warning (and probably After examination of male specimens from recognition and mating, as well) purposes Florida, we support their conclusions which and we consider it most unlikely that dis­ may be summarized by the statement of location and relocation of the antenna to Bate (1881:226) "The eyes are those of the antennular plate has occurred more the Scyllaridae; the second pair of antennae than once in the history of the family. are those of the Palinuridae. The legs are Our conclusion that the development of a common to both forms, and the carapace "stridulating line" from a nonstridulating belongs to neither. The frontal region stock is also supported by palaeontological resembles neither; and the posterior re­ evidence (given later in this paper) which sembles both, as also does the pleon, EVOLUTION OF SPINY LOBSTERS 805

whereas the pleopoda are modelled on the Palaeopalinurus Bachmeyer, 1954. Jurassic. type of those of the Scyllaridae, and the Palinustus A. Milne-Edwards, 1880. Three (pos­ sibly four) extant species in East Indies, off tail fan is that of both." We agree with East Africa, Japan, and West Indies. Bate and de Man that Palinurellus should Astacodes Bell, 1863. Lower Cretaceous of Europe. be placed in a separate family, the Syn- Justitia Holthuis, 1946. Accepting Gordon's axidae. Nevertheless, Palinurellus assumes (1960) broad concept to include Nupalirus Kubo, 1955, three named (possibly five involved special significance since several workers actually) extant species off Mauritius (two have acknowledged its primitiveness (e.g., spp.), Japan, Hawaii, and West Indies. Boas, 1882) and its morphological affinities Palinurus Fabricius, 1798. Four extant species, with fossil genera (Winkler, 1882). three in the east Atlantic, and one off East Africa. This genus has had many fossil species The review by Holthuis (1946) provides included in it but a critical examination of basic distributional data for the species them has not been attempted. of Palinuridae. This has been brought up PanuUrus White, 1847. Nineteen species in the to date by reference to unpublished data tropic and subtropic waters of the Indian, (R.W.G.) and to publications after Holthuis Pacific, and Atlantic Oceans. (1946). These include Anon (1964), Bar­ nard (1950), Bruce (1965), Edmondson Incertae Sedis (1951), Forest and Postel (1964), George Cancrinus Munster, 1839. This Liassic genus is (1962), George and Grindley (1964), regarded as intermediate between the palinurids and the scyllarids because of the club-shaped George and Holthuis (1965), Gordon antenna. (1960), Holthuis (1954, 1961, and 1963), Palinurina Munster, 1839. Liassic of Europe. Holthuis and Villalobos (1962), Holthuis Eurycarpus Schluter, 1868. Upper Cretaceous of and Zaneveld (1958), Holthuis and Loesch Germany. (1967) and Kubo (1955, 1963). The de­ scriptions given by Woods (1931) have Characters indicating phylogenetic trend. served as a basis for the fossil genera with —After separation into Silentes and Stri­ reference also to Glaessner (1960) and dentes, the extant genera have been ar­ Piveteau (1953). An estimate of the num­ ranged (Table 1) according to the degree ber of fossil species in each genus is not of development of the following four mor­ attempted here. In the Palinuridae, there phological features: (a) the relative size are eight extant and six fossil genera rec­ and disposition of the supraorbital proc­ ognized by us as follows: esses; (b) the elevation of the eyestalks; (c) the structure of the pleopod on the Silentes second abdominal segment of the female; Projasus George and Grindley, 1964. One extant (d) the general shape of the carapace. species off East Africa. Except for the last, these generic features Archaeocarabus McCoy, 1849. Nummulitic = conform to the four criteria given earlier Eocene of England, USA, and Fiji. for good phylogenetic indicators. In the Jasus Parker, 1884. Seven extant species at separate southern circumpolar localities; one Table, the size and degree of spination of Miocene species from New Zealand (Glaessner, the antennular plate in the Stridentes is 1960). also included and a trend from a narrow Stridentes to a broad structure is seen. There is Puerulus Ortmann, 1897. Four extant species a similar trend in each group towards near New Guinea, off Zanzibar, Arabian Sea, the formation, lateral separation and eleva­ and East Indies. tion of the supraorbital processes; the Linuparus White, 1847. Two extant species named (Bruce, 1965). Possibly embraces four separate elongation and elevation of the eyestalks; species, off East Africa, Japan, South China the enlargement of the endopod of the Sea, and off north eastern Australia; numerous pleopod of the female second abdominal fossil species mainly in the Cretaceous of Europe somite (see illustrations at the top of and North America. Four subgenera have been recognized (Piveteau, 1953). Figure 1) and the rounding of the carapace. 806 R. W. GEORGE AND A. R. MAIN

SILENTES STRIDENTES

PROJASUS II || PUERULUS I I JUSTITIA | | PANULIRUS I I I I II II II II 1sp 7spp PRESENT ^Spp ^Spp ^Spp 5spp ^Spp 19spp

FIG. 1. Postulated palinurid evolution. Positive fossil records are indicated black, probable records are dotted. The form of the front of the carapace and of the endopod of the female pleopod on the second abdominal segment is illustrated for the extant genera. EVOLUTION OF SPINY LOBSTERS 807

The effect of the morphological trend of analysis of the carapace furrows of the the first two structures is to provide the living and fossil decapod Crustacea. Our animal with better vision by raising the interpretation of the phylogenetic relation­ eyes and still protect them by the develop­ ships within the Palinuridae is given in ment of the forward-directed, curved Figure 1. supraorbital horns. The full expansion of The emergence of the Stridentes from the the endopod of the most anterior pleopod pemphicid stock occurred early in the fossil of the female considerably increases the record of the family since the Stridentes effective area of the whole pleopod. It ap­ Palaeopalinurus was already developed in pears that this correlates with a trend from the Jurassic; the presence of its poorly de­ deep to shallow water. It will be shown veloped supraorbital processes places it later that the genera with raised eyes and close to the Puerulus—Linuparus group in well expanded endopods (e.g. Panulirus, Table 1. Of these two genera, Linuparus Palinurus, and Jasus) live at much shal­ is well represented from the Cretaceous to lower depths than those with lowered eyes the present, suggesting that Linuparus (and and small endopods (e.g., Linuparus, probably Puerulus too) was established by Puerulus, and Pro jasus). It may be inter­ the beginning of the Cretaceous. Another preted that improved vision is an advantage fossil Stridentes which can be placed with in shallow well-lit waters and that an in­ confidence in Table 1 is the Lower Cre­ creased surface area of the pleopod pro­ taceous Astacodes] its features of sub- vides better mechanical aeration for the cylindrical carapace, moderate-sized slightly eggs in the shallow warm waters of lowered elevated supraorbital horns and a triangu­ oxygen levels. It might also be argued lar antennular plate place it close to that the less calcified carapaces of the Palinustus. As mentioned earlier, revision deeper inhabitants rely on longitudinal of the fossil forms referred to Palinurus ridging for strength and that in the shal­ is required; Piveteau (1953) records it lows, the shell is strengthened by calcifica­ with certainty in Lower Tertiary but not tion. with certainty before that time. Characters considered to be of little Fossils of two Silentes, Archaeocarabus phylogenetic importance.—In addition to and Jasus, are known. Archaeocarabus the abdominal sculpturing mentioned pre­ from the Eocene has a well marked rostrum viously, there are three features which with small pedate clasping processes; the probably are readily reversed in their evo­ carapace is cylindrical, the supraorbital lutionary development. These are: (a) the horns are moderate sized and slightly relative proportions of the legs; (b) the raised, and the eyes are probably elevated distal expansion of the propodus of the rather than horizontal. Placed in Table first walking leg to form a subchela with the 1 it would be between Projasus and Jasus, dactyl; (c) the dentition of the abdominal probably closer to Jasus. A distinguishing pleura. feature of Archaeocarabus is the well de­ It is noteworthy that these same features veloped, subchelate first leg, a feature not and their variations are often observed in obvious in the other Silentes genera. Jasus species of many other crustacean groups jlemingi (the only fossil Jasus known) has in both the living and the fossil forms. The been recorded by Glaessner (1960) from value of these characteristics for subgeneric the Miocene of New Zealand and it has identification is undoubted. close affinities to one of the present species, /. verreauxii. Thus the chronological order PALEONTOLOGICAL EVIDENCE of appearance of fossil genera in the The family Palinuridae has evolved from geological column (Figure 1) is in general the Pemphidea in the early Mesozoic, ac­ agreement with the order of arrangement cording to Glaessner (1960) following his of extant genera presented in Table 1. 808 R. W. GEORGE AND A. R. MAIN

TABLE 2. Classification of living palinurid genera based on environmental temperature.

Subpolar Temperate Subtropic Tropic Puerulus 2 spp. Puerulus 2 spp. Pro jasus 1 sp. Linuparus 2 spp. Palinustus 2 spp. Palinustus 1 sp. and 1 sp. (part) Justitia 2 spp. Justitia 1 sp. Palinurus 3 spp. Palinurus 1 sp. Jasus 6 spp. Jasus 1 sp. Panulirus 5 spp. Panulirus 14 spp. and 2 spp. (part)

GEOGRAPHICAL AND VERTICAL parkeri—south-east coast of South Africa DISTRIBUTION lat. 33°S, 620 m depth; Palinustus mos- Members of the Palinuridae live in all sambicus—Mossambique lat. 25°S, 406 the major oceans of the world. Some are m depth. Linuparus trigonus—Portuguese East Africa lat. 24°S, 360 m depth; Justitia of commercial importance and inhabit the mauritiana—off Mauritius lat. 20°S, 80 m coastal waters in tropic, subtropic, or tem­ depth. Panulirus spp. (5)—upper east perate temperature zones (see Vaughan, coast of South Africa lat. 20-0°S, 0-30 m 1940 for description of zones) while others depth; and Puerulus carinatus—Saya de are less frequently taken since they live Mahla lat. 11°S, 250 m depth. at much greater depths (down to 1400 m) where subpolar temperatures exist (Table CORRELATION OF PHYLOGENY AND 2). DISTRIBUTION In Figure 2, the distributions of extant genera have been plotted against depth Comparing Table 1 and Figure 2, the and latitude and it is apparent that, with sequences of genera show a marked cor­ few exceptions, the spatial distributions of relation which we interpret as showing genera do not overlap. Isotherms of selected that the primitive genera live in deeper, temperatures are also given in this figure, higher latitude areas whereas the advanced indicating the approximate temperature genera live in shallower warmer regions preferences of each genus. Although several in lower latitudes. genera (e.g., Palinustus, Linuparus, Pali­ The opinion expressed here of an evo­ nurus, and Jasus) all have approximately lutionary trend in the Palinuridae from the the same temperature preferences of 10-20 relatively stable conditions of the deeper C (Table 2), they in fact rarely inhabit waters to the more varied and fluctuating geographic regions which correspond in conditions of the shallow waters is in accord latitude and depth. This is well illustrated with the generally held zoogeographical by the palinurids living off the east, south, principles of Darlington (1957:547) who and lower west coasts of South Africa describes the main pattern of dispersal where every genus in the family is repre­ (for the terrestrial and freshwater animals sented by one species; on present evidence he examined) as directional, ". . . from it appears that each species occupies a the largest and most favorable areas, . . . region to the exclusion of other species of . . . , into smaller and less favorable areas." the family, thus: In the sense of Darlington, the deeper parts Jasus lalandii—west coast of South Africa of the zones inhabited by palinurids are the lat. 20-35°S, 0-50 m depth; Palinurus largest and most favorable areas, while gilchristi—south-east coast of South Africa the shallow littoral areas are less favorable. lat. 25-35°S, 60-520 m depth; Projasus For the extant palinurids, the most re- EVOLUTION OF SPINY LOBSTERS 809

FIG. 2. Distribution by depth and latitude of extant palinurid genera. Approximate isotherms are included. cent radiations into the many varied habi­ Archaeocarabus (England, North America, tats of the shallower areas are found in the and Fiji), Palaeopalinurus (Europe) and genus Panulirus (19 species). That similar Astacodes (Europe). Most of the marine radiations into relatively shallow regions fossil deposits have been described from have occurred in the past (as indicated by shallow water faunas. Little is known of fossil deposits in shallow waters) is strongly the ancient faunas which lived in the deeper suggested by the presence of many species waters since these are rare in fossil deposits. of Linuparus in the Cretaceous and early Perhaps the fossils known today represent Tertiary of Europe and North America as specialized shallow water species which died well as the more isolated occurrences of out at that time and the actual ancestors of 810 R. W. GEORGE AND A. R. MAIN the modern fauna survived in deeper, cooler "deeper water" includes the upper part of waters and were not directly represented the bathyal zone and the lower part of the in the fossil deposits. sublittoral zone and our "shallow water" covers only the upper sublittoral. Jasus DISTRIBUTION PATTERNS is regarded as a derivative of Projasus We recognize two fairly well marked emerging at or before the Miocene (/. distribution patterns for the Palinuridae; jlemingi Glaessner, 1960, is from the a high latitude, partial or complete circum- Miocene of New Zealand) and was prob­ polar distribution and a low latitude, ably well established before the most recent circumequatorial distribution. genus (Panulirus) emerged about the close High latitude distribution.—The genera of the Pliocene. An analysis of fragmenta­ which fall into this group are Projasus, tion and isolation of the circumpolar Jasus Palinurus, and Jasus (Fig. 2); these may species can not be attempted at this stage best be regarded as actual or relict repre­ of our knowledge. sentatives of circumpolar distributions. Equatorial distribution.—The remaining Jasus today has six very closely related palinurid genera occur in a broad belt ex­ species fragmented around the Southern tending for about 30° on each side of the Ocean; each species is endemic to one of equator and at depths from littoral to over the following areas: Tristan da Cunha, 700 fathoms. Eckman (1953) recognized S.W. Africa, St. Paul and Amsterdam Is., four major barriers to species distribution southern Australia, New Zealand, and Juan in this region, two terrestrial (Panama and Fernandez I. The remaining species of Suez Isthmuses) and two oceanic (Central Jasus, J. verreauxii, is quite distinct from Pacific and Atlantic); these barriers sepa­ the above /. lalandii group of species and rate four major faunal assemblages—the occurs off the North Island, New Zealand, Indo-West Pacific, East Pacific, West At­ and off New South Wales, Australia. lantic, and East Atlantic faunal regions. Only one species of Projasus (from one Recent revisions have greatly improved locality off East Africa) is known at present our knowledge of distributions and inter­ and this species may represent a survivor relationships for the shallow water genera of a genus which once had a circumpolar but the more primitive genera, living in distribution similar to that of Jasus. deeper water, are not so well studied be­ Palinurus, with all except one of its species cause of their rarity in collections. It is occurring in the north-east Atlantic, might evident in the following discussions that have once had a circumpolar distribution in endemic or suspected endemic species occur the northern hemisphere but today sur­ in the following areas: south-west Indian vives only in the Atlantic and south-west Ocean, East Indies, Japan, and West In­ Indian Oceans. dies ; these are apparently areas with a wide Both Palinurus and Projasus are gen­ range of suitable habitats. erally from greater depths than Jasus. In a) Deeper water equatorial distribution.— view of the apparent importance of depth The genera which fall into this category (and temperature) in the phylogeny of the are Puerulus, Palinustus, Justitia and palinurid genera it is convenient to treat Linuparus. the palinurids under two sub-headings: Puerulus has four closely related species shallow water and deeper water. The shal­ in the Indian and western part of the low water region extends from 1 to 50 Pacific Ocean; the species appear to be meters and the deeper waters from about geographically isolated from one another 50 to 1400 meters. This division does not and inhabit the upper regions of ocean strictly correlate with faunal zones such as basins. P. carinatus has been recorded at those proposed by Hedgpeth (1957); our Saya de Malh'a on the eastern side of the EVOLUTION OF SPINY LOBSTERS 811

Mascarene Basin, P. sewelli on the eastern no submedian spines on the anterior margin and western sides of the Arabian Basin, of the carapace. In our view, this second P. velutinus on the western side of the species group warrants subgeneric recogni­ Banda Basin, and P. angulatus on the tion and the name Nupalirus is available southern side of the Caroline Basin. Since for it. The species of Justitia occur in these species occur at fairly considerable intermediate depths (40-200 meters), the depths of 250 to 1438 meters, they are rare "Nupalirus" group occupying the deeper in collections and further collecting would part of the range. be essential to confirm their correlation Linuparus inhabits a similar depth range with particular ocean basins. (100-400 meters) to that of the "Nu­ There are three named species of palirus" group but four geographically iso­ Palinustus; these were originally described lated species are probably represented from Japan (P. waguensis), Portuguese within the recorded distribution of Linu­ East Africa (P. mossambicus), and West parus; one in Japan (trigonus), one in Indies (P. truncatus). Further specimens South China Sea (sordidus), one off Por­ from the Sulu Sea (Holthuis, 1946) prob­ tuguese East Africa (see Barnard, 1950) ably represent a new species according to and one off east Australia (see McNeill, the list of differences given by Kubo 1953; Anon, 1964). Dr. D. J. G. Griffin (1963). All Palinustus species live at shal­ (pers. comm.) has pointed out to us some lower depths (80-400 meters) than Puerulus differences between a male from 130 meters spp. off Queensland and the male described by Gordon's (1960) broad generic concept Barnard. of Justitia is represented here for discus­ b) Shallow water equatorical distribu­ sion. She united Justitia and Nupalirus tion.—Panulirus is the only genus in this (Kubo, 1955) following her examination category. The view that Panulirus is the of the holotype of /. mauritiana (Miers) most recently evolved genus of the Stridentes 1882 from Mauritius and another Mauritian has been discussed earlier. This genus is the specimen which she identified as /. japonicus largest in species numbers (19) and is (Kubo) 1955 (type locality Japan). The represented in all of Eckman's (1953) equa­ type of the genus is /. longimana (H. torial regions. Some species of Panulirus Milne-Edwards) 1837 from the West In­ have lost some or all of the exopods of the dies; Edmondson (1951) and Tinker second and third maxillipeds. According (1965) gave this name to specimens from to the degree of maxilliped modification, Hawaii. Our evaluation of the species con­ the species can be placed in groups (I, II, cerned in Gordon's broad generic concept III, IV see Table 3) and since all the suggests that two species groupings are other palinurid genera have fully formed perhaps valid. The first group has very exopods of these appendages those species long subchelate (in the male), dorsally of Panulirus with the most reduced exopods spinose first pereiopods; about four trans­ are regarded as the most advanced. In verse grooves on the abdominal terga; and Table 3, the distribution and temperature submedian spines on the anterior margin zonation of the recognized species of of the carapace—the species with these Panulirus is also indicated. The following characteristics are longimana (West In­ detailed description of the distribution of dies), mauritiana (Mauritius) and the the species in each of the faunal zones is specimens from Hawaii. The second group given to facilitate the subsequent discus­ of species, japonicus (Japan) and Gordon's sion of speciation mechanisms in Panulirus. specimen from Mauritius, have in common short non-chelate (in the male), ventrally INDO-WEST PACIFIC REGION spinose first pereiopods, about six trans­ Eleven of the 19 species of Panulirus verse grooves on the abdominal terga; and inhabit this region, and for Groups I—III 812 R. W. GEORGE AND A. R. MAIN

TABLE 3. Species groups in Panulirus based on condition of second and third maxilliped. Group I is considered to be the most primitive species group. Those species marked by asterisk live in subtropic water temperatures, the remainder are inhabitants of the tropic water temperature zone. Two species, P. longipes and P. argus, live in both tropic and subtropic zones.

Exopod condition Panulirus spp. 3rd maxilliped 2nd maxilliped Indo-West Pacific East Pacific West Atlantic East Atlantic I Present, with Present, with * japonicus argus flagellum flagellum * marginatus interruptus * pascuensis * cygnus longipes II Present, no Present, with penicillatus guttatus flagellum flagellum echinatus III Absent Present, with polyphagus laevicauda rissonii full flagellum IV Absent Present, no homarus gracilis flagellum (except ornatus where it is versicolor inflatus reduced P. homarus) stimpsoni there are equivalent morphological species In the Indo-West Pacific region it ap­ represented in other faunal regions. pears that each species has a different In the Indo-West Pacific species in general center of abundance. P. versicolor Group I, P. longipes is the only species is probably most abundant along the north with a wide distribution; there is some coast of Australia, P. ornatus is apparently evidence that subspecific recognition should more abundant at Zanzibar, P. homarus is be given to two forms, one inhabiting the the basis of a small fishery in the Arabian Indian Ocean region and the other the Gulf area (FAO/UN, 1963), P. polyphagus west Pacific region (George and Holthuis, is only recorded in any quantity on the 1965). The four remaining species each west coast of India and P. penicillatus has have restricted distributions in peripheral been reported as commercially abundant areas of the Indo-West Pacific region— P. japonicus (Japan), P. marginatus (Ha­ from the Revillagigedo and Galapagos Is­ waii), P. pascuensis (Easter and Pitcairn lands (Holthuis and Loesch, 1967). Is.) and P. cygnus (west Australia). These last four species are sufficiently abundant EAST PACIFIC REGION in their respective areas for commercial The three species along the American exploitation. west coast have adjacent distributions with The species in Groups II-IV (with the little overlap; P. interruptus inhabits the exception of P. stimpsoni which is restricted subtropic water temperature zone of the to Hong Kong) are widely distributed over California-Baja Californian region and is most of the region, particularly the Indian less frequently found in the Gulf of Cali­ Ocean. Only P. penicillatus extends east­ ward to Hawaii; this species also crosses fornia, the region which is dominated by the east Pacific oceanic barrier where it P. inflatus. P. gracilis dominates the Cen­ occurs on offshore islands adjacent to the tral American coast from the Gulf of Cali­ American coast from Galapagos to Revil- fornia to Peru. It also occurs at the lagigedo Islands. P. penicillatus has the Galapagos Is. where it is displaced from widest distribution of any palinurid and the very shallow water part of its range would be termed by Briggs (1961:545) by P. penicillatus (Holthuis and Loesch, "transpacific." 1967). EVOLUTION OF SPINY LOBSTERS 813

WEST ATLANTIC REGION the past have had to contend with the Of the four species in this region, P. changing environment and one would ex­ argus has the widest distribution, from pect that this would result in similar North Carolina and Bermuda to Rio de species distribution patterns regardless of Janeiro, and throughout this range is the group chosen for examination. For the generally the most abundant species. Two few examples we have considered, this ap­ other species, P. guttatus and P. laevicauda, pears to be so (see p. 817). occur in the central part of the P. argus Speciation in the most primitive deep range; (they are rare in the Florida-Cuba water genera.—Regarding the evolution of region and do not appear to extend to the Palinuridae, speculation on the partic­ Brazil). P. echinatus extends eastwards ular events leading to the present day from its only mainland occurrence in Brazil distribution of the older, more primitive to the islands of Cape Verde, St. Helena genera of the bathyal waters can only be and St. Peter, and St. Paul Rocks (Chace, tentative. The extant species of these 1966). In this respect, P. echinatus has genera are few in number and, within each the ability to achieve an extensive oceanic genus, all species are very similar mor­ distribution similar to that of P. penicil­ phologically. We suggest that the geologi­ latus. cal and oceanic barriers which divided the once continuous equatorial seas of the EAST ATLANTIC REGION world (including the Tethys Sea) into its Only one species, P. rissonii, is at present present day faunal regions fragmented the recognized and it is of commercial impor­ early stocks, and that the subsequent mor­ tance to French fisherman visiting the phological changes in these groups were Mauritian coast and Cape Verde Archipel­ very slow. ago. For the genus Puerulus, we have al­ DISCUSSION ready suggested that each species is as­ Darlington (1964) expresses the view sociated with a separate ocean basin; per­ that the continents of Africa and South' haps each species has been evolved in America were probably joined during the response to a set of environmental con­ Palaeozoic but that separation occurred ditions peculiar to each1 of these basin in the early Mesozoic or late Palaeozoic. regions. One could imagine such a situa­ In our consideration of palinurid evolu­ tion operating after the initial, larger scale, tion, we are primarily concerned with the barriers were erected. The peculiarity of post-Mesozoic when the positions of the the fauna in other ocean regions which oceans and land masses were probably much have very obvious topographical barriers as they are today. In these oceans we can (such as the Red Sea, Mediterranean, envisage distribution patterns of marine Philippine Basin, and the Japan Sea) is animals which differ little from those of well known (Sverdrup et al., 1961). Just the present day. Naturally the species may which environmental factors are likely to be different. Habitat zones of the oceans limit the distribution of the species of were probably filled by species which re­ Puerulus are, at our present state of acted in much the same way as do their knowledge, obscure; perhaps the answer present occupants. There is little reason lies in a better understanding of the sub­ to imagine habitats remaining vacant for surface circulation of these and adjacent any length of time; following extinction ocean basins. of a specialized group within a habitat as Speciation in Panulirus.—Panulirus in­ a result of some environmental deteriora­ habits sublittoral waters and during the tion, the habitat would be filled from a Pleistocene ice ages this genus would be new source fairly rapidly. The reactions subject to great environmental stresses, e.g., and interactions of all marine species in falling and rising sea levels, changes in 814 R. W. GEORGE AND A. R. MAIN area of continental shelves, changes in past is of paramount importance and the water temperatures, and strengthening and following sets out our interpretation of the weakening of oceanic circulation. nature of these changes and their probable A consideration of the pelagic larval effects on Panulirus speciation. Oceanic stages (phyllosoma) in the speciation of circulation is driven by the wind systems Panulirus is important since the larvae of the globe and during the advances and have great potential for wide dispersal. retreats of ice in the Pleistocene the circula­ However, dispersal does not appear to be tion is believed to have strengthened and haphazard. Information on larval develop­ weakened. When the wind circulation is ment, larval behavior, and the dispersal weak, oceanic currents are probably weak and recruitment paths of the larvae in and diffuse; pelagic larvae contained in response to current circulation are known these currents would be widely dispersed only for a few species of Panulirus. Nothing and settling of larvae would not be con­ is known about the phyllosoma of all the tained within limited geographical areas. more primitive genera. Of other genera When wind circulation is strong, oceanic for which phyllosoma are recognized, only currents are probably well defined and very few species have all stages described. persistent; in these conditions strong cir­ Fewer species still have adequate field data culation is likely to result in the fragmenta­ of sufficient phyllosoma stages (associated tion of an earlier widely-distributed adult with information on local current systems) stock into discrete populations which are to allow the presentation of possible re­ unconnected by larval dispersal. cruitment paths. Nevertheless, the results Perhaps the four species groups of of some field studies on the biology of Panulirus indicate the effects of four major commercial species of Panulirus suggest to Pleistocene glaciations, each glaciation pro­ us that larvae are not entirely randomly ducing conditions for optimum isolation dispersed but remain reasonably confined and subsequent speciation. In each of three to the general area of adult distribution; species groups in Panulirus (Group I, II, this is based on the findings of Johnson and III in Table 3), analagous species to­ (1960#, b) for Panulirus interruptus off the day occur in two or more separate faunal coast of California and of George and regions; if these groups reflect close genetic Cawthorn (1962) and Chittleborough affinity, and if four species groupings are (1964) for Panulirus cygnus off the west the result of previous Pleistocene glacia­ coast of Australia. In these studies, the tions, then it is reasonable to assume that larval stages are shown to gradually move there have been equatorial connections be­ offshore but mechanisms which confine and tween the Pacific and Atlantic Oceans in return the later stages to the onshore fish­ recent times to allow such genetic inter­ ing ground are as yet unknown. Whether change. Following Simpson's (1952) postu- the offshore late stages represent the lation that dispersal and interchange of "wasted" portion of the larvae or the terrestrial faunas does not necessarily in­ "true" recruits remains to be shown. But dicate entire and persistent migration path­ the larvae are rarely found beyond the ways during the particular geological period latitudinal geographical limits of the coastal under discussion, the evidence of post adult population; this must surely be the Tertiary dispersal of terrestrial faunas be­ result of behavioral responses (e.g., by tween North and South America does not, vertical migrations) of the larvae within by any means, rule out the possibility of prevailing current systems. intermittent marine communications be­ If the species of Panulirus have evolved tween the Pacific and Atlantic oceans dur­ in response to water circulation patterns, ing the Pleistocene. a consideration of what might have hap­ Recent marine workers such as Briggs pened to these systems in the geologic (1961), Brinton (1962), and Davies EVOLUTION OF SPINY LOBSTERS 815

I ndowcst pacific cast west east I Pacific Atlantic lAtlantic

Pliocene

1st Phase

2nd Phase

3rd Phase

4 th Phase

Ul Present 3 Ul uda Species 3 n o Ul o c .^c: 3 vie o CD H~>

FIG. 3. Postulated speciation of Panulirus.

(1963), have apparently assumed an un­ Panulirus may have evolved before the broken Panamanian Isthmus during the Pliocene and altered little since that time. Pleistocene and have interpreted the recent In these latter two alternatives, the evolution of marine species as resulting Panulirus groupings would have no genetic from changing sea temperatures. There is significance and their similarities would be no doubt that temperature and other en­ simply convergent. vironmental changes have played a very The following hypothesis for speciation important role in marine evolution but in Panulirus assumes that there were four we believe that considerations of the changes opportunities for marine interchange in in total current systems and the possibility the Panamanian region. The postulated of marine interchange in the Panamanian evolution in Panulirus is represented in region during the Pleistocene are more Figure 3. The ancestral species was able significant than considerations of tempera­ to inhabit the sublittoral of an extensive ture alone. If there were no marine con­ geographic area in the equatorial region nections after the late Pliocene, one must during the late Pliocene with free access envisage parallel speciation in Panulirus to all ocean areas and was "protected" from on each side of an unbroken Panamanian potential extinction by its wide environ­ Isthmus with a similar parallel sequence mental tolerance in both the larval and evolving in the Indo-West Pacific with adult condition. This tolerance would per­ this procedure occurring three times. As mit wide dispersal of larvae over a range an alternative, all the present species of of temperatures (probably subtropic and 816 R. W. GEORGE AND A. R. MAIN tropic) and habitats. The modern species in flatus. P. homarus might best be ex­ which fit closest to this description are plained as a development from P. penicil­ P. longipes and P. argus; we have chosen latus in particular, because of the presence to represent "the" ancestral species by P. of the small flagellum on its second maxil- longipes because it is better known to us. liped exopod. By the end of the first glaciation, one An examination of the habitat prefer­ could imagine the Indo-West Pacific species ences of each of the Indo-West Pacific —P. japonicus, P. marginatus, P. pascuensis, species indicates that (a) where closely and P. cygnus—as peripheral, isolated rep­ related species occupy a similar habitat, resentatives of "P. longipes" each species they are found to be geographically isolated perhaps responding to separate "endemic" (japonicus, Japan; marginatus, Hawaii; current systems. In the east Pacific and in pascuensis, Easter I., and cygnus, Western the west Atlantic, P. interruptus and P. Australia, all live in rock and coral at sub- argus respectively were isolated from one tropic temperatures); and (b) where the another by a temporary Panamanian bar­ tropic species are sympatric over a very rier, and from the Indo-West Pacific forms wide area, each appears to be separated by by the Pacific Ocean barrier (Fig. 3). habitat preference. For instance at Ceylon During the succeeding interglacial, full where all six tropical species occur, De oceanic interchange of species is again Bruin (1962) showed that each species available due to diffuse current circulation has preferential habitat, depth and be­ and the breaking down of equatorial land havior. The different combinations of these bridges. Following the re-establishment of characteristics for each species presumably the oceanic and land barriers of the next determine which one predominates in a glaciation, populations which earlier passed particular area; De Bruin's work in Ceylon through the broken "barrier regions" would clearly illustrates such regional domination. have been isolated, and one could imagine Comparisons outside the Palinuridae.— P. penicillatus, P. guttatus, and P. echinatus Comparisons can be made between the pat­ as further derivatives of any one of the terns of distribution of the Palinuridae and earlier species—P. longipes, P. interruptus of other groups of marine animals as long or P. argus. Following similar reasoning, as one bears in mind that groups being at the end of the next glaciation P. compared must have been classified ac­ polyphagus, P. in flatus, P. gracilis, and P. cording to similar sets of taxonomic values. laevicauda probably originated from any Where the evolutionary relationships of a one of the species P. penicillatus, P. gut- group have been worked out, such an ap­ tatus, and P. echinatus. Panulirus rissonii praisal can be used to compare such taxo­ originated from either one of the west nomic levels. It would be surprising if Atlantic species of the previous glaciation distribution patterns of other groups were or from an earlier stock which came from absolutely identical with those outlined P. argus at the first glaciation. here for the Palinuridae, but since the The most recent separation of the re­ various marine groups of animals have maining Panulirus species, all of which oc­ been subjected to the same overall environ­ cur in the Indo-West Pacific, probably mental stresses in the past, similarities are developed in response to temporary iso­ to be expected. Factors such as the age lation within restricted parts of this vast of the group, their habits, dispersal poten­ region—suggestions of which may be in­ tial (as larvae and adults) modify the dicated by the various areas of relative patterns of distribution. Although there are abundance of each species witnessed today. numerous examples of similar distribution P. ornatus, P. versicolor, and P. stimpsoni patterns for a wide variety of marine ani­ might have originated from any one of the mals, our remarks here are restricted to species P. penicillatus, P. gracilis, and P. some of the decapod Crustacea—the group EVOLUTION OF SPINY LOBSTERS 817 with which we are most familiar in this taxonomic groups. For instance in the respect. In the decapod crustacean groups Indo-West Pacific Ocypode (beach crabs) examined, Eckman's (1953) concept of the species distribution pattern is very four major shallow water, circumtropical similar to that shown by Panulirus; some faunal regions was closely supported as are widespread (O. ceratophthalma and O. was the generally accepted thesis that the cordimana) whereas others form a group warm parts of the oceans usually contain of peripheral endemics in Japan, Hawaii, a larger number of species than the cool western Australia, and Madagascar (com­ waters. parable with the peripheral distribution Where sufficient reliable taxonomic in­ pattern of the Panulirus japonicus species formation for the group was available to group). There are, however, more Ocypode make a valid comparison with the palinu- species, compared with Panulirus, living as rids, the distribution patterns and affinities endemics within relatively small geographi­ were found to be consistent with the palinu­ cal areas in the tropical region of the Indo- rid patterns, e.g., the evolution of the West Pacific region; tropical endemics of Portunidae and the distribution patterns Ocypode are known from South Arabia, of Ovalipes (Portunidae) and lbacus (Scyl- west coast of India, east coast of India, laridae). The general pattern of evolution Borneo, north coast of Australia, and from deeper to shallower water is clearly Coral Sea. illustrated in the Portunidae. Stephenson Background information such as this has (1962) assessed the affinities of the portu- assisted us in our approach to a "difficult" nid subfamilies and he regarded the Portu- species, O. kuhlii. It was orginally de­ ninae as evolving from the more primitive scribed from Java and many subsequent Carcininae and Macropopinae. Our analy­ workers have stated that it occurs at-many sis of the depth and geographic distribu­ widely separated localities in the Indo- tion of these crabs shows that the few West Pacific region. Recent carcinological species that inhabit depths greater than studies in Western Australia (George and about 100 fathoms or those living in high Knott, 1964) show that previous records latitudes belong to the two primitive sub­ of O. kuhlii from west and north Australia families. All the species of the derivative aremisidentifications; Crosnier (1965) also Portuninae are inhabitants of warm shallow showed that previous records of O. kuhlii waters. The southern circumpolar distribu­ from Madagascar are very doubtful. These tion pattern of closely related species, results suggest that O. kuhlii is another of typified in the Palinuridae by Jasus, is the tropical endemics inhabiting only a shown by several decapod genera such as limited region (around Java) and is not Ibacus and Ovalipes (this genus is cur­ a widely distributed species. We might also rently being revised by Prof. W. Stephen­ venture to suggest that the proliferation of son, pers. comm.). In both these examples, species of Ocypode has been in response however, at least one species of the genus to a proliferation of different habitats along also occurs in subtropic Japanese waters the shore during the Pleistocene and that (but no species occupy the intermediate the species probably evolved from a per­ tropic region). A combination of two dis­ manently aquatic sublittoral ancestor which tribution patterns (southern circumpolar in turn evolved from a deeper water an­ and peripheral Indo-West Pacific) is thus cestor. evident for these two genera. Origin of marine faunas.—It is well to Where the taxonomy of a group is not note that we do not suggest by extrapola­ well known, we have sometimes found it tion that the abyssal or hadal fauna is useful to make comparisons with our broad ancestral to the bathyal fauna which in palinurid conclusions to assist in the un­ turn is ancestral to the sublittoral and ravelling of "difficult" species in these supralittoral faunas. On the contrary, we 818 R. W. GEORGE AND A. R. MAIN

support the suggestions by other authors (and also Pro jasus and Palinurus in relict (e.g., Brunn, 1956) that the abyssal and form) and a circumequatorial low latitude hadal fauna evolved from the shallower pattern; we have made a further sub­ bathyal fauna. However, one must be division into the shallow water species cautious of the interpretation that the (Panulirus spp.) and deep water inhabi­ bathyal fauna might have come from the tants (the remaining genera). The species warmest, shallowest water. For considera­ of one of these deep water genera—Puerulus tion, we suggest that the intermediate —appear to be associated with oceanic (bathyal) depths may have been the origi­ basins. nal area and that radiations have proceeded 6) The warmest shallowest regions con­ up into shallower waters (the crustacean tain the most number of species because of situation discussed in this paper) as well the greater diversity of habitat but these as down into deeper waters (the suggestion species are the most vulnerable in a chang­ by Brunn, 1956 and others). ing total environment. 7) Speciation within the most recently SUMMARY evolved palinurid genus—Panulirus—has 1) Spiny lobsters occur in all major occurred during the Pleistocene and the oceans of the world from sublittoral to four species groups probably reflect four bathyal depths. A phylogeny of the Pali- major glaciation periods. Interglacials al­ nuridae is proposed based on external mor­ lowed equatorial inter-ocean dispersal of phology. The family is divisible into two the pelagic larvae, while during glacials groups—Silentes and Stridentes and within stocks were isolated by physical barriers each group a parallel trend in several mor­ and more defined current systems. phological features is apparent. 8) It is suggested that the bathyal 2) Coincident with these morphological region for many marine organisms is the trends is a habitat trend from deep original region and that radiations have (bathyal) to shallow (littoral) waters; proceeded both up into shallow water and adaptive advantages under these conditions down into the deeper abyssal regions. are considered. 3) The shallow waters are the special­ ACKNOWLEDGMENTS ized habitats of spiny lobsters and the fos­ We wish to formally acknowledge the sils indicate past radiations into shallow willing assistance given by our many col­ water but not the surviving stock; it is leagues from Universities, Museums and suggested that the "successful" ancestral other Institutions in the compilation of line is represented by the deeper water material and literature for this work. We inhabitants. are also particularly grateful to Professor 4) An analysis of depth and latitude H. Waring and Dr. D. Bradshaw of the distributions of the palinurid genera shows Zoology Department, University of West­ that the spatial distributions of the genera ern Australia, Drs. W. D. L. Ride, G. M. rarely overlap; in other words, each genus Storr, and B. R. Wilson of the Western appears to have a separate set of depth Australian Museum and Dr. W. G. Inglis and latitude parameters. In the southwest of the British Museum for their very useful Indian Ocean one species of every palinurid criticisms in the final stages of the prepara­ genus is represented presumably because tion of the manuscript. the total region provides all the separate habitats for these genera. LITERATURE CITED 5) There are two basic distribution pat­ ANON. 1964. Prawn trawler nets rare "Barking terns in the Palinuridae, a circumpolar high crayfish." Fish. News Lett. Aust. 23(11): 26. latitude pattern, as exemplified by Jasus BARNARD, K. H. 1950. Descriptive catalogue of EVOLUTION OF SPINY LOBSTERS 819

South African decapod Crustacea. Ann. S. Afr. ern Australian crayfish; in Western Fisheries Mus. 38: 1-864. Research Committee, Fisheries Department, BATE, C. S. 1881. On Synaxes, a new genus of Western Australia (Mimeo.). Crustacea. Ann. Mag. Nat. Hist. (ser. 5) 7: GEORGE, R. W., AND J. R. GRINDLEY. 1964. 220-228. Projasus—a new generic name for Parker's BOAS, J. E. V. 1882. Die Guttung Synaxes Sp. crayfish, Jasus parkeri Stebbing (Palinuridae: Bate. Zool. Anz. 1882: 111-114. "Silentes"). J. Roy. Soc. W. Aust. 47: 87-90. BRIGGS, J. C. 1961. The east Pacific barrier and GEORGE, R. W., AND L. B. HOLTHUIS. 1965. A the distribution of marine shore fishes. Evo­ revision of the Indo-west Pacific spiny lobster lution 15: 545-554. of the Panulirus japonicus group. Zool. Verh., BRINTON, E. 1962. The distribution of Pacific Leiden 72: 1-36. euphausids. Bull. Scripps Inst. Oceanogr. 8: GEORGE, R. W., AND MARY E. KNOTT. 1964. 51-270. The Ocypode ghost crabs of Western Australia BRUCE, A. J. 1965. A new species of the genus (Crustacea: Brachyura) J. roy. Soc. W. Aust. Linuparus White, from the South China Sea 48: 15-21. (Crustacea Decapoda) Zool. Meded., Leiden GLAESSNER, M. F. 1960. The fossil decapod 41: 1-13. Crustacea of New Zealand and the evolution BRUNN, A. FR. 1956. The abyssal fauna: its of the order Decapoda. Palaeont. Bull. N.Z., ecology, distribution and origin. Nature, Lond. no. 31: 1-63. 177: 1105-1108. GORDON, I. 1960. On the genus Justitia Holthuis CHACE, F. A. 1966. Decapod crustaceans from Decapoda, Palinuridae). Crustaceana 1: 295- St. Helena Island, South Atlantic. Proc. U.S. 306. Nat. Mus. 118: 622-662. HEDGPETH, J. W. 1957. Classification of marine CHITTLEBOROUGH, R. G. 1964. Western Aus­ environments. In: Treatise on marine ecology tralian crayfish-recruitment phase; in Western and paleoecology. Mem. Geol. Soc. Amer. 67: Fisheries Research Group, Half-yearly Report 18-28. January-June 1964, Fisheries Department, HOLTHUIS, L. B. 1946. The Decapoda Macrura Western Australia (Mimeo.). of the Snellius Expedition. Temminckia 7: CROSNIER, A. 1965. Crustaces decapodes. Grap- 1-178. ridae et Ocypodidae. Faune Madagascar 18: . 1954. On a collection of decapod Crustacea 1-143. from the Republic of El Salvador (Central DARLINGTON, P. J. 1957. Zoogeography: the America). Zool. Ver. 23. geographical distribution of animals. Wiley, . 1961. The taxonomic status of Panulirus New York, 675 p. echinatus Smith 1869 (Decapod Macrura, . 1964. Drifting continents and late Paleo­ Palinuridae). Crustaceana 2: 223-227. zoic geography. Proc. Acad. Nat. Sci. Philad. . 1963. Preliminary descriptions of some 52: 1084-1091. new species of Palinuridea (Crustacea, Deca­ DAVIES, J. L. 1963. The antitropical factor in poda, Macrura, Reptantia). Proc. Acad. Sci. cetacean speciation. Evolution 17: 107-116. Amst., (ser. c) 66: 54-60. DE BRUIN, G. H. P. 1962. Spiny Lobsters of HOLTHUIS, L. B., AND A. VILLALOBOS. 1962. Ceylon. Fish. Res. Stat. Dept. of Fish. Ceylon. Panulirus gracilis Streets y Panulirus injlatus Bull. No. 14: 1-28. (Bouvier), dos especies de langosta (Crustacea, ECKMAN, S. 1953. Zoogeography of the sea. Decapoda) de la costa del Pacifico de America. William Clowes, London. An. Inst. Biol. Univ. Mex. 32: 251-276. EDMONDSON, C. H. 1951. Some central Pacific HOLTHUIS, L. B., AND J. S. ZANEVELD. 1958. De crustaceans. Occ. Pap. Bishop Mus. 20: kreeften van de Nederlandse Antillen. Zool. 183-243. Bijdr. no. 3: 1-26. FAO/UN. 1963. Report to the Government of HOLTHUIS, L. B., AND H. LOESCH. 1967. The Aden on the Crawfish Resources of Eastern lobsters of the Galapagos Islands (Decapoda, Aden Protectorate. Based on the work of R. Palinuridea). Crustaceana 12: 214-222. W. George. Rep. FAO/EPTA, (1696) 1-23. JOHNSON, M. W. 1960a. The offshore drift of FOREST, J., AND E. POSTEL. 1964. Sur une espece larvae of the Californian spiny lobster nouvelle de langouste des iles du Cap Vert, Panulirus interruptus. Symposium. California Palinurus charlestoni sp. nov. Bull. Mus. Hist. Cooperative Oceanic Fisheries Investigations Nat. Paris (ser. 2) 36: 100-121. Report no. 17: 147-161. GEORGE, R. W. 1962. Description of Panulirus . 1960&. Production and distribution of cygnus sp. nov., the commercial crayfish (or larvae of the spiny lobster, Panulirus inter­ spiny lobster) of Western Australia. J. Roy. ruptus (Randall) with records on P. gracilis Soc. W. Aust. 45: 100-110. Streets. Bull. Scripps Inst. Oceanogr. 7: GEORGE, R. W., AND P. CAWTHORN. 1962. In­ 413-462. vestigations on phyllosoma larvae of the West­ KUBO, I. 1955. Systematic studies on the Japa- 820 R. W. GEORGE AND A. R. MAIN

nese macrurous decapod Crustacea. 5. A new South Atlantic with special reference to the palinurid, Nupalirus japonicus, gen. et sp. no v. Mesozoic. Bull. Amer. Mus. Nat. Hist. 99: J. Tokyo Coll. Fish. 41: 185-188. 81-258. . 1963. Systematic studies on the Japanese STEPHENSON, W. 1962. Evolution and ecology macrurous decapod Crustacea. 6. A new and of portunid crabs, with special reference to imperfectly known species of palinurid lobster. Australian species, p. 311-327. In: G. W. J. Tokyo Coll. Fish. 49: 63-71. Leeper [ed.]. The evolution of living orga­ MCNEILL, F. 1953. Carcinological notes, no. 2. nisms. Melb. University Pr., Melbourne. Rec. Aust. Mus. 23: 89-96. SVERDRUP, H. W., M. W. JOHNSON, AND R. H. MAN, J. G. DE. 1881. Carcinological studies in FLEMING. 1961. The Oceans. Prentice Hall, the Leyden Museum, No. 1. Notes Leyden New York. Mus. 3: 121-144. TINKER, S. W. 1965. Pacific Crustacea. Tuttle . 1882. On the genus Araeosternus de M. Company, Tokyo. Notes Leyden Mus. 4: 161, 162. VAUGHAN, W. T. 1940. Ecology of modern PARKER, J. J. 1884. On the structure of the head marine organisms with reference to paleo- in Palinurus with especial reference to the geography. Bull. Geol. Soc. Amer. 51: 433. classification of the genus. Trans. N.Z. Inst. WINKLER, T. C. 1882. Carcinological investiga­ 16: 297-307. tions on the genera Pemphix, Glyphea and PIVETEAU, J. (ed.). 1953. Traite de Paleontologie. Araeosternus. Ann. Mag. Nat. Hist. (ser. 5) Masson, Paris. 3. 10: 133-149, 306-317. SIMPSON, G. G. 1952. Probabilities of dispersal WOODS, H. 1931. A monograph of the fossil in geologic time, p. 163. In E. Mayr [ed.]. macrurous Crustacea of England. Palaentol. The problem of land connections across the Soc, London.