Publication of Such a Monumental Survey in 10 Volumes Over a Span Of

REVIEWS

EDITED BY WALTER BOCK

Handbook of the Birds of India and Pakistan, together with those of Bangladesh, Nepal, Sikkim, Bhutan and Sri Lanka.--Salim Ali and S. Dillon Ripley. 1971-1976. Bombay, London, and New York,Oxford UniversityPress.--The "Handbookof the birdsof India and Pakistan"by Ali and Ripley is a major 10-volumeornithological review of the fauna of the great Indian subcontinent.The first four volumeswere reviewedby JoeMarshall (Auk 87: 816-818, 1970;89: 207-209, 1972)to which the readeris referred for a discussionof the project'sscope and an evaluation. The presentreview coversthe last six volumes, which treat all the passerinesexcept the pittas, which are in volume 4. Instead of evaluatingthe passerinematerial in detail, I shall presentan overview of the whole series. Publicationof sucha monumentalsurvey in 10 volumesover a spanof 8 yearsresults in a numberof technicalpoints that shouldbe mentioned.In thesetimes of rapidlychanging politics and national events, noteven the title of an ornithologicalwork issafe, as shown by thechange in thesubtitle of thishandbook. Originallythe subtitle read "together with those of Nepal,Sikkim, Bhutan and Ceylon" but with the publicationof volume8 (1973), it read "togetherwith thoseof Bangladesh,Nepal, Sikkim, Bhutan and Sri Lanka." And it shouldbe noted that the territorial conceptof Pakistan in the main title is not the sameat the endof the seriesas in the beginning.The authorspresent (vol. 8: xiv and subsequentvolumes) a list of changesin territorial limits and names. Altering in the handbook title could cause citation problems.I recommendciting the whole series,including the first sevenvolumes, by the latter title. A secondproblem is the year of publication,because the date in severalvolumes does not alwayscoincide with the informationmade available to me by the publisher;these dates are listed below with the date printedin the volumegiven in parenthesesif it differsfrom that providedby the publisher.I wouldalso like to usethis opportunity to providethe mostrecent price of all ten volumesas follows (the earlier dates are those of publication in India, the later onesin America, citationsshould use the earlier dates). 1, Divers to hawks, 380 pp, 1968, out of print• 2, Megapodesto Crab Plover, 345 pp, 1969, $20.00• 3, Stone curlews to owls, 325 pp, 1970 (1969), $19.75 4, Frogmouthsto pittas, 265 pp, 1970, $21.75 5, Larks to the Gray Hypocolius, 276 pp, 1972, $17.50 6, Cuckoo-shrikesto babaxes, 245 pp, 1971, $20.00 7, Laughing thrushesto the Mangrove Whistler, 236 pp, 1973 (1972), $17.00 8, Warblers to redstarts, 277 pp, 1973, $21.75 9, Robins to wagtails, 306 pp, 1974 (1973), $20.00 10, Flowerpeckers to buntings, 344 pp, 1976 (1974), $20.00 The "Handbook of the Birds of India and Pakistan" is a monumental and masterful analysis of the avifauna of the Indian subcontinent,covering a total of 2,060 speciesand subspecies.The importanceof the project is underlined by the decreasingnumber of ornithologistswith a breadth of knowledge of natural history, systematics,and distribution who could undertake such a project in the future. This decreaseof ornithologistspossessing a broad overview of a large avifauna has been demonstratedclearly to me by my failure to find a competentreviewer of this handbook;I surelyam not. During my entire tenure as Review Editor, I have tried without successto find a reviewer for this work, and have beenforced to undertake the task myself becauseof its importance.Hopefully with the help of someexpert advice, I will be able to do justice to these volumes. The entire avifauna of the Indian subcontinenthas beencovered in this handbookusing a goodup-to-date taxonomy for the speciesand subspecies.A total of nearly 2,100 forms (speciesand subspecies)of some 1,200 speciesare covered.Each form is given a referencenumber that correspondsto the number in Ripley's"A synopsisof thebirds of India andPakistan" (1961). • All speciesand subspecies are cross-indexed with Baker's "Fauna of British India, birds." Two new speciesare added in the addendato volume 10, the Rednecked Grebe (Podicepsgriseigena) in volume 1 and the Egyptian Nightjar (Caprimulgusaegyptius) in volume 4. The descriptionsare good, but it is sometimesdifficult to obtain a visual picure from the text description.

• Accordingto Dr. Ripley vol. 1 and hopefullyvol. 2 will be reissuedby Oxford University Presswith additions and corrections.He has alsobrought his Synopsisof 1961 up to date for republicationin Bombay, hopefully within a year. 791 792 Reviews [Auk, Vol. 94

Color illustrations of all speciesare included. Many of the color plates are from previouslypublished books, someof which are out of print; the rest of the plates were prepared especiallyfor this handbook. Someraces that are illustrated are not found in the area coveredby thesevolumes which can causesome difficulties. Not all of thesespecies are shownin plates closeto the text discussion,and someare far out of place. Two species of swallows are illustrated in plate 108 (of a total of 113)in volume 10; swallowsare treated in volume 5. Yet when one turns to volume 5, the correctreference to plate 108 is given showingthat the format of the entire set of 10 handbookvolumes, including the plates, was carefully planned at the onsetof the project. Each bird on the color plates is identified by its scientificand English namesand with its referencenumber. The plates are attractive and generally quite good. Unfortunately the sourcesof the reprinted plates are not given; this information would have been useful to non-Indian specialistssuch as myself. The keys are good and work well in general;but onemust identify the bird to family or subfamilybefore the keyscan be used. Salim All has contributeda vast amount of previouslyunpublished natural history data which constitutes an important contribution of thesevolumes. Moreover an excellentjob was donesummarizing the literature and unearthinginformation from sourcessuch as Whistler'sunpublished notes. Distribution, status, and habitat have been updated as far as is known. The well-presentedmaps are a very useful feature of this handbook. Unfortunately Mercator projection maps of the Eurasian land masswere used, which deempha- size the area the handbookcovers. A detailed map of the handbook area is not included nor is a gazetteerof place names. Some place names mentioned are no longer in use. And in some casesthe distribution is described from district to district without saying in which state the districts are. The last six volumes on the passerinebirds have better speciesdescriptions than those given in the first four volumes. It is possibleto get a goodfeel for the plumage pattern of the bird. Volume 10 containsan eclecticbibliography of the 744 titles on the birds of the Indian subcontinentup to the end of 1972. It also contains a consolidated index of scientific names and one of common names: My review copy of volume 10 is lacking pp. 311-314--the centersheet of the signature•of the index, the only major error of printing that I have found. Most of the shortcomingsof thesevolumes have beencovered by Marshall in his reviews of the first four volumes and need not be repeated here. The "Handbook of the Birds of India and Pakistan" can be summarized as a superbsummary of the birds of the Indian subcontinent.The authorsgathered, summarized, and compiledthe knowledgeof the status, distribution, habitat, and life historyof thesebirds and presentedit clearlyand concisely.They attempteda limited review of the subspeciesof the birds of the handbook region. The major criticism is that the authors attemptedfew new thingseven when they couldhave been readily done, suchas remeasuringan adequate sampleof each taxon and presentingthe resultsin a more completeway (samplesize, mean, and standard deviation in addition to the range). The original choiceof format with each subspeciestreated as a separate unit and the inclusionof colorplates resulted in a muchlarger and moreexpensive work, well beyondthe financial abilities of most individuals and many libraries. This 10 volumes costingalmost $200 could have been reduced to one-half or even one-third of its size without reduction in content and quality with a different schemeof organization.Yet thesecriticisms fade away in the light of the fact that Ali and Ripley have madeavailable to us a thoroughand excellentaccount of a major avifauna of the world. Moreoverthey completedthe publicationof the 10 volumeswithin a period of 8 years, so that the last volume appeared before the information in the first volume became stale and out of date. They presented a summary of this avifaunathat recordsthe knowledgeof the distributionand habitat of thesebirds before the habitatand the birds are completelydisrupted. The "Handbook" will remain the standardreference to the "birds of India and Pakistan,together with thoseof Bangladesh,Nepal, Sikkim, Bhutanand Sri Lanka" for manydecades. All ornithologistsare indebted to Salim All and S. Dillon Ripley and their coworkersfor this excellent treatise. The Bombay Natural History Societyand Oxford University Pressare to be congratulatedfor their roles in sponsoringand publishing it.--WALTER J. BOCK.

Ecology and evolution of commnnities.--Martin L. Cody and JaredM. Diamond (Eds.). 1975. Cambridge,Belknap Press of Harvard UniversityPress. Pp. ix + 545, illus. $35.00.--RobertMacArthur had a profoundinfluence on the developmentof modernecology and populationbiology, fostering especially theemergence of communityecology from a descriptiveto an analyticscience. His ideasabout avian ecology were the cornerstoneof the broaderapplications of histheories and richlywarranted the CouesAward that was conferredby the AmericanOrnithologists' Union in 1974. A year after MacArthur's tragic death in November1973, his colleagues met at Princetonin a memorialsymposium. The papersin thisvolume grew out of that gatheringand portray the spirit and diversity of approachesand questionsthat MacArthur inspired in his associates. October1977] Reviews 793

The 18 chapters are organized into 3 sections.The first considersaspects of the diversity and relative abundanceof species,how thesemay be analyzed, and how stable the relationshipsmay be. The following sectiondeals with the strategiesemployed by coexistingspecies for the division of limiting resources.The final, longestsection details the waysin which communitiesare structured,how this structuremay develop, and how the organization is determined by forces of competition and predation. The two concluding chaptersdeal in a broader manner with MacArthur's contributionsto ecologyand look to the ways in which his ideas may develop or be applied in the future. Altogether, the contributionsprovide a broad but somewhat disjointed synthesisof where MacArthur's ideas have led his followers and how they may influence the further development of community ecology. In his 1972 book, GeographicalEcology (Harper and Row, New York), MacArthur observedthat "To do scienceis to search for repeated patterns, not simply to accumulate facts, and to do the scienceof geographicalecology is to searchfor patternsof plant and animal life that canbe put on a map." MacArthur's work exemplified the formative stagesof this search for and understanding of patterns, and the contributionsin this volume indicate how far his studentsand colleagueshave taken it. There are papersdealing with plants (Schaffer and Gadgil, Horn), butterflies (Shapiro), stream plankton (Patrick), lizards (Pianka), birds (Hespenheide,Cody, Karr and James,Diamond), mammals(Rosenzweig, Brown), and a little of everything (Hutchinson). Some contributionsclarify existing models of speciespatterns; May's chapter, for example, provides a detailed assessmentof speciesabundance models and attempts to determine which patterns are truly elementsof the biological structuring of communitiesand which are merely statistical consequences. Other papers(e.g.J. MacArthur) deal with well-known patterns, such as the broadly defined latitudinal gradientsin speciesdiversity, in the light of additionalenvironmental variables (in this case,winter-summer differencesin mean temperature). Still other chaptersdevelop new ways of revealing and perhaps understanding the patterns of communities.Much is made, for example, of Levins' developmentof loop analysis, which provides a mathematical framework for portraying the logical and somenot-so-logical consequences of speciesinteractions, and the ways in which these may contribute to community evolution. Several of the papers deal at length with bird communitiesand thus merit specialcomment here. A good deal of current theory holds that regular size differencesbetween closelyrelated coexistingpredators imply correspondingdifferences in the sizesof prey taken, thereby permitting coexistencein the face of potential competition.Hespenheide addresses this dogma, usingrather undetailedinformation on the dietsand body sizesof several major groups of birds. There are indeed patterns, but they are subject to a variety of influences:differences in foraging behavior, prey type preferences,and the extent of intraspecificsocial aggregationmay all modify the body size/preysize relationship. Studiesthat attempt to infer ecological relationshipsdirectly from morphologicalmeasures are thus subjectto the caveatthat "it all depends."Karr and James also convincingly demonstratethe hazards of developing ecologicalgeneralizations from mea- suresof single morphologicalfeatures, but in an entirely different manner. Using the multivariate tools of principal componentsanalysis and canonicalcorrelation to considersuites of ecologicaland morphological traits at once, they discernpatterns of convergencein the overall eco-morphologicalconfigurations of forest avifaunas in Panama, Liberia, and Illinois. Unfortunately, the ecologicalcategories are of necessityquite broad, and the morphologicalanalyses consider only measurementsrecorded from a single male specimen for each species,so the details of communityorganization and populationvariation cannot be evaluated. Cody's paper, prompted by the premise that similar forces of natural selectionin similar but separated environmentsshould produce similar optimal solutionsof communitystructure, considers in somedetail the patternsof speciesdiversity on severalgeographical scales, emphasizing the breedingbird assemblagesof Mediterranean climate habitats in California, Chile, and Africa. The paper is packed with inventive analysesof his observations(which are tabulated in several appendices).These lead to the conclusionthat localwithin-habitat diversity can be explainedadequately by theoriesof competitionand limiting similarity of species,while differencesin regionalor between-habitatdiversities among the threecontinents reflect the influencesof history and chance. Cody treats the accumulation of speciesnumbers with increasingarea and habitat diversity on continentsin a truly innovative fashion, but the treatments(and the conclusions)are marred by defectsin the basic observations.Avian community compositionwas gauged by a seriesof short-termsurveys at many locations,using small (4-10-acre) censusareas; the durationand spatialscope of the surveysmay have been too short and too small to censusaccurately the actual breedingassemblages, although the most abundant specieswere undoubtedlyencountered. This inadequacyof the censusesis apparently compensatedfor by including in the tabulations speciesthat were "expected"to occurin a census but did not do so. The main thrust of the analysisconsiders the distributional spreadand turnover of species along habitat gradientsin the three continents,but where most of the occurrencesof a speciesalong a gradientare by "expectation"rather than actualsightings (e.g. 3 of 4 censuslocations forA mphispizabelli in 794 Reviews [Auk, Vol. 94

California, 6 of 7 for Parisoma subcaruleumin Africa, and 13 of 45 speciesin a "supersaturated"California habitat), one becomesuneasy about the reality of the patterns. The habitat gradientsare scaledby featuresof vegetationprofile, which operationallyseems to carry the assumptionthat all important factorsin determin- ing avian community patterns are incorporated in these measures. Diamond's massive(102-page) chapter presentsthe details of avian community compositionon New Guinea and its many satelliteislands, and detectsseveral "assembly rules" for avian communities.These are dictatedby resourcecompetition: "Communities are assembledthrough selection of colonists,adjustment of their abundances,and compressionof their niches, in part so as to match the combined resourceconsump- tion curve of all the coloniststo the resourceproduction curve of the island. Members of permitted combinationsmust also be 'companionsin starvation'--i.e. must be similarin their tendenciesto overexploit and in their tolerancesfor loweredresource levels, thereby starving less tolerant species off the island.Thus consumerspecies form hierarchieswith respectto exploitativestrategy." These are important and contro- versialideas; unfortunately there seems no ready way to testthem exceptin highly fragmented,well-studied archipelagos.Diamond interprets his findings in the context of diffuse competition,the web-like net of competitive interactionsamong many membersof communities.Diffuse competition, however, is even more difficult to demonstratethan direct competitionand seemsto be usedas an explanationof community patterns when interpretationsbased on simple direct competitiontheory fail. The characteristicsof the "MacArthur School" approach are apparent in most of the contributions. Interpretationsof patternsare frequentlymade in an evolutionarycontext, with referenceto adaptive costs and benefits.The observationsthat providethe fodderfor theoreticaldevelopment are usuallyderived from short-term(typically one year) surveysof areas,rather than from experimentalmanipulations or long-term intensive studies. We are often asked to acceptthe observationsas valid samplesof natural community organization without knowing the methodsor circumstancesof their collection. The emphasisis seemingly on the analysisof the data and on generatingintuitive or mathematicalmodels that generallyagree with the observations.In most treatments,competition is viewed as the major force driving the developmentof community patterns (the papers of Patrick and of Connell are notable exceptions),although the role of competition is only inferred from the patterns of differencesamong coexistingspecies or from the distributional patterns of presumedcompetitors. The natural communitiesare nearly always presumedto be at or near equlibrium, but givenonly single-yearsurveys, there is really no way to tell whetherthis assumption really holds. A massive and at times truly elegant and novel complex of analysesis thus built upon a data base that is gatheredin a rather cursoryway and upon assumptionsthat are untestedbut questionable. Despitethese apparent inadequacies of the approach,broad patternsin communityorganization do emerge, and the analyseslead to some interesting interpretations of the patterns, at least some of which can be formulated as testablehypotheses. MacArthur, after all, was interestedin principlesthat held generally, that could explain the major portion of the variable patterns of nature. Perhaps intensive, detailed, long-term and completestudies of local communitiesare not really requiredto reveal the broad outlinesof these patterns, and if that is so the contributionsin this volume have a great deal to say about the organizationof communities.But MacArthur alsobelieved that this sort of ecologyis perhapsbest pursued by field naturalistswho know theory and can placetheir observationsin that context.As one readsthrough this compendiumof where the "MacArthur School"stands now, it is hard to avoid the feeling that the balanceof theory with natural history advocatedby MacArthur has swungtoward theory and away from natural history.--JOHN A. WIENS.

Island biology illustrated by the land birds of Jamaica.--David Lack. 1976.Vol. 3, Studiesin ecology.Berkeley, University of California Press.xvi + 445 pp. $25.00.--This book, written by David Lack shortlybefore his death in 1973, is basedon 10 monthsof intensivefield work on Jamaicaduring 1970 and 1971. It is, in part, a detailedaccount of the ecologyof Jamaicanbirds. But Lack alsouses Island Biology to state his views on the origin and evolution of island faunas, views reflectingmore than three decadesof observationon, and thinkingabout, islandbirds. This bookis especiallyimportant because it provides an alternative to prevalent theoriesconcerning island biogeography. Lack's theme is that an island's avifauna reflects the heterogeneityof its environmentsand not its geographicalisolation. According to Lack, taxa dispersereadily enougheven to remoteislands so that the ecologicalspace on islandsis filled to the samedegree as it is on continents.To supportthis idea, Lack gatheredevidence both from his ecologicalstudies on Jamaica and other islands,and from the geographic distributionsof specieswithin island groups.This evidencemakes up the first half oflsland Biology. The secondhalf is a detailed,species-by-species account of theecology of Jamaican birds. At the endof the book 29 appendices summarize Lack's observations. October 1977] Reviews 795

BecauseIsland Biology has so many facets, I shall use a correspondingnumber of perspectivesin this review, examiningthe bookas (1) a descriptionof the ecologyof Jamaicanbirds, (2) an expositionof Lack's theory on the faunal diversity of islands, (3) an alternative to other theoriesof island biogeographyand ecology,and (4) a philosophicalattitude toward science,which statesthat natural systemscan be understood better by detailedobservation of their structureand function(a 'close'view) than by testingpredictions resultingfrom hypothesesabout their organization(a 'distant' view). Lack beginswith an expositionof his theoryof islandbiogeography. Next, he describesJamaica and its habitats, the manner of ecologicalsegregation between the land birds, the history of the avifauna, and geographicaldisplacement of specieswithin the West Indies. The remaining chaptersin part 1 deal with peripheral topicsincluding comparisonsbetween Jamaica and other islands, the origin of the speciesof Jamaicanbirds, and the ecologyof wintering warblers.Part 2 describesthe ecologyand systematicsof each speciesbelonging to the familiesof birds from pigeonsthrough passerines.The 29 appendicesof tabulated data includefeeding observations, counts of speciesin varioushabitats, and geographicaldistribution. The book is concludedby a rather thorough list of referencesand indices to birds and islands, but not, unfortunately,to ideasand concepts.For the mostpartIsland Biologyis well written, but Lack sometimes repeatshis pointsso often that they losetheir impact, and he frequently listsso many examplesas to make his narrative tedious and numbing. Consideringthe diversityand uniquenessof the Jamaicanavifauna, it hasbeen the subjectof surprisingly few ecologicalstudies. Lack's bibliographylists only eight paperson Jamaican birds publishedsince 1950. But in spiteof the opportunitiesthat Jamaicaoffers the ecologistand evolutionarybiologist, Lack's account will be of little interestto the ornithologistwho does not specializein theJamaican fauna; the bookalso is not suitable for the casualbirdwatcher on Jamaica. It would seem that Lack wrote Island Biology more for himself than for a generalaudience. It is overly long for its accomplishmentsand Lack's ideasmight have been more stronglypresented and more widely read if they had been publishedin a scientificjournal. Simply stated, Lack's theory of island biologyis that the habitats on an island determinethe number of specieson the island, with distancefrom the sourceof colonistsplaying a decidedlyminor role. Lack also believedpopulations of birds on islandsto be stable;once a populationis established,its extinctionand replacementby a new colonistis unlikely. How strongis Lack's evidencefor theseideas? On the whole, it appears quite weak. To support his view that dispersal within the West Indies is not restricted by water barriers, Lack citedobservations on movementsof birds betweenislands and providednumerous anecdotal recordsof Caribbeanbirds as vagrants outside their normalranges. Such occurrences, often of a singlebird during the nonbreedingseason, may have little relevanceto the establishmentof populations.If one can generalizefrom speciescurrently invading the West Indiesfrom SouthAmerica--the Bare-eyedThrush and the Glossy Cowbird--colonizationfollows widespreadmovements by large numbersof birds. Many individuals usuallyare seenbefore breedingpopulations are establishedand colonizationproceeds in a steadyprogression from island to island rather than the haphazardfashion in which vagrantsappear. Lack himselfacknowledged the obviouseffect of distanceon the avifauna of suchremote islands as Hawaii, but apparently wasunwilling to concedean effectof smallerscale in the much closerCaribhean islands. Lack's opinionis all themore amazing considering that severalfamilies of mainlandspecies, including most of the suboscinepasserines, are absentfrom the West Indies where suitable habitats almost surely exist. Discountingthe effectof isolation,Lack explainedvariation in the numberof specieson islands,and the generalpoverty of islandfaunas compared to thosefound on continents,by the reducedvariety of habitats availableto birds--ecologicalimpoverishment. The relationbetween habitat diversityand faunal diversity is evidentamong islands with low profilesand thereforelacking wet, montanehabitats. But Jamaicarises to an altitude of over 2,000 m and its habitats range from extensivewet forestsat mid elevationsto more restrictedxeric scrubs in the lowlands.In his Table 2, Lack distinguished13 terrestrial habitats on Jamaica; there are 66 speciesof resident land birds. Lack made an extensivecomparison of his data with Monroe's (1968, Ornithol. Monogr. No. 7) studyof land birdsin Honduras,where 389 speciesoccupied a recognized 10habitats, but he presentedno evidence of Jamaica being ecologically impoverished. For a theorythat rests sostrongly on the ecologicaldiversity of a region,the absenceof adequatehabitat comparisons is disappoint- ing. Lack alsoignored the influenceof within-habitatheterogeneity, a functionof the diversityand structure of the vegetation,on the numberof speciesfound in uniform habitats. There is no questionthat eachisland habitat has fewer speciesof birds than its mainland counterpart,yet it is doubtful that the structure of the island habitatsis lesscomplex in any way that is meaningfulto birds (Cox and Ricklefs 1977, Oikos 28: 113-122). Anotherfacet of Lack'sview of islandbiology concerns competitive exclusion among close relatives. If one assumed,as Lack did, that habitatswere fully saturatedon islands,one would predictthat interspecific competition should be as intense on islands as it is on the mainland. To support this idea, Lack cited 796 Reviews [Auk, Vol. 94 examplesof geographicalsegregation of closelyrelated (ecologically similar) species within the West Indies. But, althoughLack documentsmany interestingcases of adjacent,but nonoverlappingdistributions, his surveyis neither thoroughnor systematic.Only thosecases that supportthe theory are emphasized;the manyinstances of gapsbetween the geographicalranges of similarspecies and the many casesof generic sympatryare playeddown (13 generaof Jamaicanland birdsare representedby morethan onespecies). It is easy to dismissthe contrary evidencesuggested by sympatric congenersusing the obviouspoints of divergencebetween them as a convenientout. But is it impossibleto distinguishwhether thesedifferences are inherentand permitthe speciesto coexistor whetherthey are the resultof co-occurrenceand subsequent divergent evolution. Competition underlies another of Lack's ideas about islands, namely that, once established,island populationsof birds effectivelyexclude new colonistshaving similar ecologicalrequirements. Although islandresidents certainly may adapt to the peculiarconditions on an island, they typicallyoccupy greater ecologicalranges than their mainland counterparts.There is no theory of competitionthat, in general, would permit the regularexclusion of specialistsby generalists,particularly in supposedlystable island environments.I do not presumeto know the origin of Lack's idea concerningthe stability of the island avifauna, but it is similarto Williams's(1969, Quant. Rev. Biol. 44: 345-389) stronglysupported view of a similar nature on the anoline lizards of the West Indies. If residentsexcluded new colonists,there would be little extinctionand faunal turnover on islands.To supporthis contentionthat island avifaunas are stable, Lack pointed out that during the last 150 years-- correspondingto the periodof detailedavifaunal investigations--only two specieshave becomeextinct on Jamaica,both morethan a centuryago, and nonehas colonized the islandwithout the help of man, despite the frequentappearance on Jamaicaof vagrantsfrom otherislands and the mainland. Perhapsbecause he concentratedhis attention on Jamaica, Lack overlookedthe many casesof extinctionthroughout the West Indies(33 islandpopulations of land birdshave eitherbecome extinct in the past 150years or are certainto becomeextinct soon (Bond 1968, 13thsuppl. to the check-listof the birdsof the WestIndies, Acad. Nat. Sci. Philadelphia)and severalcases of recentgeographical expansion within the LesserAntilles (Bond 1956, Check-listof thebirds of the WestIndies, 4th ed., Acad.Nat. Sci.Philadelphia). One could argue that most of these extinctionand colonizationevents are the result of human disturbance,although this would be difficult to supportin severalcases. More to the point, 150years of recordedornithological history are hardly sufficientto make a casefor faunal stability on an islandthat hasbeen inhabited by birdsfor tensof millions of years. Lack'sviews contrast strikingly with otherhypotheses concerning island biogeography. In the prefaceto his book, Lack suggestedthat the differencesbetween his theory and that of MacArthur and Wilson are less than he had oncethought; Orians (1976, Science194:1153-1154) alsoplayed down differencesbetween the two theories. I find them difficult to reconcile. MacArthur and Wilson stated that the faunal diversity of an islandis the resultof a balancebetween immigration and extinctionof species.To be sure,many aspects of Lack's hypothesiscan be accommodatedby this equilibrium theory. Lack believedthat birds disperseeasily through the West Indies, in other words, that the immigration curvesof the MacArthur-Wilson modelare similar for all islands. Differences between islands would reflect different extinction curves. These are determinedboth by stochasticevents and by competition,and are thought to increaseas island size and ecologicalheterogeneity decrease. Fine, sofar, but evenif variation in faunal diversitydepended upon the ecologyof the islandand not its geographicsetting, MacArthur and Wilson'smodel would predictfaunal turnover whereLack arguedfor competitiveexclusion of immigrantsby establishedresidents. If coloniza- tion rates were uniform throughout the West Indies, depauperateislands should experiencemore rapid faunal turnover with each taxon having a shorteraverage life expectancy.The absenceof distinctive endemicand presumablyolder taxa on small, depauperateislands (Mayr 1965, Science150: 1587-1588, Ricklefsand Cox 1972,Amer. Nat. 106:195-219)supports the MacArthur-Wilsonview and weighsheavily againstLack's theory of fauna] stability. Lack's argumentfor uniform colonizationrates or, rather, completelack of contemporarycolonization within the West Indies stimulated me to attempt to sort out the effectsof distanceand island characteristics on speciesdiversity using a multiple regression-partialcorrelation analysis of the data in Appendix 10 of Island Biology. The dependentvariable was the numberof speciesof residentland birds;the independent variableswere island area (transformed to its logarithm),altitude, distance from mainland,and distance to nearestisland. The first two independentvariables should provide an index to the variety and area of habitatson an island;the last two shouldprovide a measureof geographicalisolation. In partial correlation, the effectof eachvariable on the dependentvariable is evaluatedwith the otherindependent variables held constant statistically, so correlationsamong independent variables do not confound the analysis. The results,presented in Table 1, indicatethat within the Greater Antillesisland area is the only factorthat bears October 1977] Reviews 797

TABLE 1 MULTIPLE REGRESSION-PARTIAL CORRELATION ANALYSIS OF FACTORS INFLUENCING SPECIES DIVERSITY ON WEST INDIAN ISLANDS 1

Variable

Sample Distance to Distance to Multiple Islands2 size Area Altitude mainland nearest island R 2

West Indies 38 0.75 a 0.31 -0.27 -0.10 0.88 Greater Antilles 12 0.91 a 0.07 0.30 0.02 0.96 Lesser Antilles 21 0.61 a 0.56 a -0.78 a -0.46 a 0.87

i Analyzed with UCLA Biomedical Computer Program BMD 03R. e The five Virgin Islandsare not includedin either the Greater or LesserAntilles samples. a p < 0.05. any relationship to number of species.Distance and, surprisingly, elevation are inconsequential.The analysissuggests, therefore, that within the GreaterAntilles colonizationrates are uniform. It shouldbe pointed out that, among the Greater Antilles, island area and elevation are stronglyrelated (correlation coefficient, r = 0.80) and island area varies over almost four orders of magnitude. In the LesserAntilles, where island area varies over only two ordersof magnitude,all four independent variables are related significantlyto number of species,with distanceto the mainland having the highest partial correlationcoefficient (r = -0.78). The effectof isolationis evidenthere, with diversitydecreasing by 2.2 -+ 0.4 speciesfor every 100 km distancefrom South America and by 4.3 -+ 2.1 speciesfor every 100 km distance from the nearest neighboring island. Clearly, dispersal is important. A topic not dealt with explicitlyby Lack, but worth mentioningin the contextof his book,is the originof the Antillean avifauna and the sourceof its distinctiveness.Bond (1963, Proc. Acad. Nat. Sci. Philadelphia 115: 79-98) suggestedthat endemicAntillean forms are relictsof populationsthat were onceextensively distributedon the mainland--that islandsare isolated backwashes, relatively untouched by the mainstream of evolution on the continents.Lack supposed,in contrast, that island endemicswere derived from mainlandforms by adaptationto the particular conditionsthey encounteredon the islands.Species that now differ greatly from their mainland ancestorsoccupy much different habitats, while thosespecies that are undifferentiateddid not experiencesuch habitat change when colonizingthe islands,even though they may be as old as endemicforms. Ricklefs and Cox (op. cit.) arguedthat the degreeof distinctivenessof an island form reflectsthe ageof the taxonon the island:endemic forms are old colonists,undifferentiated forms more recentimmigrants. These authors additionally side with MacArthur and Wilsonin arguingfor extinction and faunal turnover. The contrastingviews of the origin of the Antillean avifauna held by Bond, Lack, and Ricklefs and Cox grew out of consideringthe samebody of factsabout the geographicaldistribution and taxonomicdiffer- entiation of island forms. Each of theseviews either explicitly or implicitly reflectsthe authors'particular conceptsof evolution,diversification, ecological interaction, and extinction.Clearly, we cannotrely on a fossil record, which is negligiblefor birds, to resolvesuch differences.And I doubt that the kind of geographicand taxonomicdata that has led to so many incompatibletheories could eventually, through more detailed analysis, be used to test these theories. Lack has arguedthat oneway to circumvent the biasesthat color our interpretation of patternsin nature is to studythe systemin greaterdetail. This, in Williams(op. cit.) words,is the 'closeview,' which, Lack says, "dependson the intensivestudy of a particular situationin all its aspects,from which conclusionsare later drawn." I certainly agreewith the value of suchclose views of nature. But while readingIsland Biology I was struck repeatedlyby Lack's failure to heedhis own warning to seekersof 'distant'views, that "because nature is sovarious, onecan with diligencefind examplesto illustratealmost any new idea." I was saddened to find thatIsland Biology is frequently little more than an attempt to marshall weak evidencebehind an outdatedconcept of islandbiogeography. There is undoubtedlymuch truth in what Lack has said about islands, but in spite of his own intensive studieson the birds of Jamaica, supportor refutation of Lack's hypotheseswill be the burden of a youngergeneration.--RoBERT E. RICKLEFS.

Bird $ound$.--Gerhard A. Thielcke. 1976.Ann Arbor, University of Michigan Press.viii+ 190 pp. ,95 black-and-whitefigs. and sonagrams.Cloth. $6.95; paper $2.95 .--In this work Dr. Thielcke'spurpose is to provide a bridge of communicationbetween bioacoustics and other sciencesand to show the importanceof avian vocalizations both as a means of communication and as a factor in evolution. Well-suited for the task, 798 Reviews [Auk, Vol. 94

Thielckehas broad experiencein bioacoustics,especially of Europeanpasserines, and an obviousen- thusiasmfor the subject. The result is a readable, conciseaccount of avian soundsand their role in the lives of birds, and one that doesnot overlap other recentaccounts on the subject. After an initialchapter on description and interpretation of vocalizationsby sonagramsand oscillograms, Thielckediscusses the typesof mechanicaland vocal sounds and the anatomyand physiology of thesyrinx and the avian ear. Most of the bookdeals with variousaspects of song.Topics include: function of song, speciesand individual recognition of song,role of learningin song,ontogeny of song,and the role of songas an evolutionaryisolating mechanism. A comprehensivechapter on typesand functions of callsis noteworthy asthis is a topicusually neglected in bioacousticalworks. I foundthe discussionon soundparasitism with examplesof parasitismby viduineson Africangrassfinches especially interesting. The bookconcludes with a brief discussionon the relationshipbetween bird songand human music. Most informationfor thisbook comes from Europeansources, including Thielcke's own publishedand unpublishedworks. Also no referencesmore recentthan 1968 are included,because Bird Soundsis a translationof Vogelstimmen(Berlin-Heidelberg-New York, Springer-Verlag),printed in 1970.Neither of thesefacts greatly detracts from the valueof the bookor its contents,as most of the principlesdiscussed still holdtrue. Few typographicalerrors are present, and thetranslation from the Germanis good.However the bioacousticalterminology, especially in thefirst chapter,is confusedby jargon.This is lessThielcke's fault than an indicationthat bioacousticsstill lack universaldefinitions. The developmentof a standardized terminology would be a helpful contribution for the advancement of bioacousticalresearch. Bird Soundsis an excellentbuy. I recommendit to all interestedin avian sounds.For the amateur, student, and scientistin nonbioacousticalfields, it will serve as an informativeintroduction to bird vocalizations.For studentsof bioacoustics,its chiefvalue will be for its literatureand insightinto Thielcke's own research.--DENNIS m. FORSYTHE.

Bird hazards to aircraft.--Hans Blokpoel.1976. Buffalo, Books Canada Inc. xiv + 236pp., 41 figs., 4 tables, and 11appendices. $6.50 (paper)and $10.00 (cloth).--Even thoughthe first recordedaircraft crash andfatality caused by abird occurredin 1912,the advent of high speed jet aircraftand a dramaticincrease in the number of aircraft flying made the seriousproblem of bird/aircraft collisionsthat Blokpoeladdresses in his book. His statedaim was to gatherthe bestinformation available on the subjectso that it couldbe appliedto reducesignificantly the hazardsbirds pose to aviation.In thistask he hassucceeded admirably, but we must wait to seeif the application of the information will reduce the severity of the problem. The bookis divided into seven chapters. Chapter 1 presents general information on birds together with an overview of bird migration, local flights, flocking, and behavior of birds with respectto approaching aircraft. Most of the informationwill not be new to ornithologists,because this chapterwas intended to provide background material for those readers unfamiliar with birds. The referencesfor this and the other chaptersare complete only through the springof 1974,and some of thestatements in thischapter are already out-of-date(e.g. birdsare not sensitive to ultravioletlight, andonly a few birdshave a senseof smelland it is poorlydeveloped). The use of Pettingill(1970, Ornithologyin laboratoryand field), Thomson(1964, A dictionaryof birds), and the old edition(1966, Fundamentalsof ornithology)of Van Tyne and Bergeras major sourcesfor the chapter probably accountsfor this misinformation. Chapter2 discussesthe natureof the collisionsbetween birds and aircraft(bird strikes)and reportsand analyzesbird strike statistics.Because the detectionand reportingof bird strikesrequires many tedious stepsand voluntary cooperationfor success,the statisticstend to be extremelyconservative. In caseswhere seriousstrikes have occurred reports are usually filed. The datafor commercialairlines are particularly hard to get for rather obviousreasons. The photographsin this chapterprovide vivid illustrationsof the typesof damagethat can resultfrom bird strikes.The informationon the costsof bird/aircraft collisionsis staggering (e.g. the U.S. Air Force alone sustainsan averageloss of $10.9 million per year, and in 1973the costsof repairingbasic structural damage to planesfrom collisionswith birds, not includingengine damage or salaries,was $24 million).It shouldbe stressedthat theseare extremelyconservative cost estimates. Chapter 2 closeswith an interestingsurvey of the bird speciesinvolved in collisionswith aircraft throughoutthe world. The chapter effectivelydemonstrates the seriousnessand global nature of the problem. The emphasisof Chapter3 is on techniquesfor makingairplanes more resistantto bird impacts,i.e. bird-proofingthe aircraftand engines.Much of the materialpresented in the chapterrelates to the design and constructionof aircraftcomponents and should prove educational to biologists.The main airworthiness requirementsregarding bird impactsare outlinedwith referenceto turbineengines, windshields, and tail structuresfor both civil and military aircraft. Stricterbird strike requirementsare clearlyneeded, but October1977] Reviews 799

engineeringconstraints argue against completely bird-proofing an aircraft. One alternativeis to investigate the feasibilityof usingon-board devices that would dispersebirds in front of the aircraft; this approachis the subject of Chapter 4. Some of the techniquesinvestigated thus far include on-board lights, lasers, and microwaves, but apparently no effective airborne deviceswill be available in the near future. The next two chaptersare perhapsthe mostuseful in the book for ornithologistswho are consultedabout bird/aircraftcollision problems. Chapter 5 dealswith the preventionof bird strikesat airports,and Chapter 6 concentrateson the preventionof bird strikesaway from airports. Chapter 5 coversbird observation methods,bird dispersalmethods, bird removaland killing methods,habitat manipulation,and planningof new airports. Blokpoelrightfully emphasizesthat bird control at airportsshould ideally begin when a new airport is planned, but for existing airports scaring and habitat alteration are the most useful means of correctinga bird problem. Trapping and killing are last resort measuresand often ineffective. Chapter 6 reviewsthe methodsfollowed to give pilotsand air traffic controllersuseful information on bird hazards duringflight. It stressesprocedures to avoidbirds in flight and discussesin detailmost of the recentwork on bird migration in terms of maps of routes, forecastmodels for intensebird migrations, and the detection, quantification, and monitoringof hazardousbird movementswith radar. As with so many endeavorsthe quality of the data baseis limiting, and this is particularlytrue when forecastsof bird migrationmust be basedon weather forecasts.Despite theselimitations, migration forecastshave been remarkably accurate and useful, particularlyin military air operations.The discussionsof radar techniquesin the chapterare good, but the statement that bird heights cannot be determined reliably with surveillance radars is misleading.Modern weather surveillanceradars (WSR) routinely measurethe height of thunderstormsand rain showers and can successfullymeasure the altitudinal distribution of migrating birds. The final chapteris onlyfour and a half pageslong and givesan overviewof the nationaland international organizationsworking on the bird strikehazard; detailed information on the nationalgroups and individuals concernedwith working on bird strike problemsin 1974is in an appendix.Blokpoel saysthat Canada, the United States,and many Europeancountries now have nationalcommittees that approachthe bird hazard problems with varying degreesof determinationand success.I find that the Inter-Agency Bird Hazard Committee of the United Statesoperates with little determinationand has accomplishedso little that its successcannot be evaluated.In contrast,Canadian and European effortshave shownmany achievements with marked success. The book contains445 references,and theseare in generalwell chosen.Some, however, are governmental agencytechnical reports and hard to obtain. The organizationof the text is suchthat any chaptermay be read and understoodwithout readingthe precedingchapters. The proofreadingwas thorough;the text is practicallyfree of typographicalerrors. Unfortunately the English systemof measurementis usedthrough- out the book, becauseaccording to Blokpoel the systemis still being used widely in English-speaking countriesand in aviation around the world. Nonethelessornithologists are not accustomedto reading about Cliff Swallowsweighing 0.046 lb and 6.3-oz ScreechOwls. The omissionof scientificnames of the numerousbirds, mammals, and plants mentionedin the text is particularly bothersome.Hopefully these shortcomingswill be correctedin later editions. Blokpoel is to be congratulatedfor compilinga great amount of scatteredinformation on bird hazardsto aircraft and presentingit in a highly readablestyle. The bookis well worth its price and shouldprove useful to a wide spectrum of readers from professionalornithologists and aeronautical engineersto airport managersand air traffic controllers.--SIDNE¾A. GAIJTHREAIJX,JR.

The Golden-cheeked Warbler.--Warren M. Pulich. 1976. Austin, Texas Parks and Wildl. Dept. xv + 172pp. Illus. byAnneMarie Pulich.Paper. 50g handlingcharge.--A rare andthreatenedbirdattracts particular interest among biologistsand conservationists.The Golden-cheekedWarbler (Dendroica chrysoparia)is sucha species,and this first detailedand comprehensiveaccount of it is assureda wide and attentive audience. The Golden-cheekedWarbler nestsonly in the Edwards Plateau and adjacent areasof central Texas (31 counties).Its nestingrequirements are met only by densestands of mature Ashejuniper (Juniperusashei) occurringin scatteredtracts known locally as cedarbrakes. The bird'sreliance on this tree apparentlystems from nest constructionwith bark stripsprovided uniquely by mature trees of this species.It winters in a comparativelysmall area of montanepine-oak forest from east-centralGuatemala through Honduras into northernNicaragua. In migration it traversesMexico, but, contraryto publishedaccounts including the A. O. U. Check-list,it probablydoes not winter there. Somerecords of occurrenceare broughtinto doubtby confusionwith otheryellow-headed warblers, particularly the Black-throatedGreen (Dendroica virens) and Hermit (Dendroica occidentalis). 800 Reviews [Auk, Vol. 94

The Golden-cheekedWarbler is not immediately in danger, but the long-rangehabitat trends give cause for concern. The total area covered by mature standsof Ashe juniper has shrunk 20% in little more than a decade, partly as a result of agricultural agencies'encouragement of cedar eradication to increasepasture land. Alsoin someplaces habitat hassteadily deteriorated because of urban sprawl. However therehas been a hearteninggrowth of local interest in the bird, aided by Pulich, and this has led to the settingaside of several tracts of prime habitat on public and private lands. Several circumstancesinvite comparisonwith the Kirtland's Warbler (Dendroica kirtlandii), which is Michigan'sown bird in the samesense that thisis Texas'sown bird. Both are restrictedduring nestingseason to small regions and distinctive habitats associatedwith one dominant kind of tree; both migrate to comparativelysmall winter rangesin the northerntropics, the Golden-cheekedin Central Americaand the Kirtland's in the Bahama Islands of the West Indies; both normally produce only one brood per year; both seemto haveno specializedfood requirement; and bothare heavilyparasitized by Brown-headedCowbirds (Molothrus ater). On the other hand, there are notable differences.The habitat of the Golden-cheeked Warbler is climax forest,conferring a degreeof permanenceto land preservesunlike the transientforest stage occupiedby the Kirtland's Warbler. The area utilized at presentin Texas (about 350,000 acres)is vastly larger than that in Michigan (about 6,000 acres),and the populationof the Golden-cheeked(15,000-17,000) is very much larger than that of the Kirtland's (400). The Golden-cheekedis the more difficult of the two to census,and estimatesof the population have been basedon appraisal of the terrain and the probable density of nestingbirds in "excellent,average, and marginal" habitats;whereas, singing male Kirtland's Warblers have been counted individually. An interestingcircumstance that might passunnoticed is that neither of thesespecies seems to fill up the habitat availableto it, althoughthe conventionalassumption is that the populationsare limited by nesting habitat. Even in the mostsuitable spots, the occupiedterritories do not embraceall the spaceavailable. For example, Golden-cheekedterritories have a mean size of 4.29 acres, but Pulich found only one pair per 20 acresin "excellent"habitat. Similarly, Kirtland'sWarbler t•rritories average8.4 acres,but the densityof birds in the best places is seldom greater than one pair per 30 acres. The relationshipwith the cowbird is intriguing. The ratio of young Golden-cheekedWarblers fledgedin Pulich'sprincipal studygroup, 27% of eggslaid, was smallerthan for any other speciesof warbler reported to date. Nearly 60% of thesenests received cowbird eggs,and the other nestsnot enteredby cowbirds producednearly all the warblersfledged. Yet the cowbird is known to have parasitizedGolden-cheeked Warblers at least as far back as the 1870s.However this account doesnot explore whether the cowbird has increasedgreatly herein moderntimes as in many other parts of its range. The samplesin this studyare not large enoughto answerthe question,but they arouseour curiosity.Of 15 setsof eggsPulich collectedin 1962-1964, 11 includedcowbird eggs;of 15 setsan oologistcollected before 1937, only 2 includedcowbird eggs. Generally the range of topics is extensive,encompassing the historical record, population, range, migration,habitat, territory, pair formation,nest building, eggs,incubation, nesting success, food habits, song,weights, and measurements.The bibliographylistsnot only literature citedbut alsocollateral reading. Pulich put tremendouseffort over a period of more than 15 years into detailed searchesby auto and on foot, as well as by visitsto collectionsand libraries, to ascertainthe speciespast and presentbreeding range, migratorypath, and winteringground. More than 50 pagesare devotedto this phaseof the study.Although not all of this will be of interestto everyreader, it putson recorddetails that will form a basefor appraising changesin the future. The data on nestingactivity and successare limited mainly to threeyears, 1962-1964, and the samplesare small. Hopefully, asPulich points out, otherswill be encouragedto do further fieldworkboth onthe nesting ground and on the wintering ground, where very little is known about the bird. This report reminds us by contrasthow much knowledgeof most bird specieshas come through slow accumulationfrom many hands. Although Pulich graciouslyacknowledges many sourcesof help, it is apparentthis work from conceptionthrough publication was mainly the productof one man and his family. He wasaided in the field by hisson, Warren, Jr., and the coloredfrontispiece and line drawingsthroughout the book were done by his wife. This will remain the standard work on the Golden-cheekedWarbler for a long time, but it is not so exhaustiveas to discouragestudy by others.This is not an easyspecies for research.Many partsof its range are difficult to reach, the cover is dense,the bird's songis not persistentnor doesit carry far, the nestsare hard to find, and the warblersare not easyto capturefor marking. Soit is not surprisingthat the samplesof data are small and in someinstances need considerable enlargement for refined analysis.For example, the samplesof weightsand measurementsof birds comparedfrom northernand southernparts of the nesting October1977] Reviews 801 rangeare oftentoo small for calculationof usefulmeans and standarddeviations, and Pulichmight have beenbetter advised to groupall examplesof eachsex from the entirenesting range for measurementof centraltendency and variance; anyway, geographical variation in a mobilecreature with sucha smallrange would be astonishing. No onecan complain about the price of thisuseful book about the Golden-cheeked Warbler .--HAROLD F. MAYFIELD.

A new guide to the birds of Hong Kong.--MichaelWebster, illustrated by KarenPhillipps, with additionalplates by RaymondWong, Takeshi Shiota, and Alex Olsson. 1976. Hong Kong, Sino-American Publ. Co. (114 How Ming St., Kwun Tong, Kowloon,Hong Kong). 111 pp., 8 color platesand 40 black-and-whiteplates. Paper. U.S. $5.95.--Thisvolume is a very usefulguide to field identification, describingconcisely and illustrating well most of the365 species of birdsreliably recorded in HongKong as well as discussinga number of other speciesthat are likely to occur. The text accountsare brief,3-12 linesper species,containing length (in inches),description and field identificationincluding habits and often habitat, status,and seasonalpresence. The descriptionsare succinctand well done, giving a goodpicture of thebird. Their brevity, though, will precludeidentifying somemore difficult species, such as raptorsthat requiremore detail; e.g. the distinctiveimmature of Spilornis cheela is not mentioned. Theillustrations are accurately and attractively done. Over 350 species are illustrated, 82 in colorand the restin blackand white. The species are unfortunately not identified by name on the plates, but by the species numberin thetext. Many of thepaintings and the drawings appear to be modeled after those in otherbooks, a procedurethat is no doubt necessary in an areathat lacks a goodspecimen collection. I am not sure how I feelabout this, but perhapsthe author should, at least,give credit to the artistwho painted the original. Becausefield guide users depend heavily on the illustrations, many (especially beginners) will havetrouble identif,vingdifficult groups such as shorebirds, warblers, and flycatchers by using the line drawings. These arefine for illustratingbirds with strikinglydifferent and contrasting patterns, but do not lend themselves to showingsubtle differences. Hopefully the publisher will allowmore color plates in futureeditions; the color platesand line drawings are designed so that new color plates can simply replace the current line drawings. The four-pageintroduction is a briefdescription of whatthe bookis all aboutand givessome tips for birding.Its mostuseful feature is a list of goodplaces to seebirds and directionsfor contactingthe Hong KongBird Watching Society. Hong Kong is one of the few places along the east coast of Asia easily accessible to bird watchers,and visitors will find thisbook helpful in gettingacquainted with its birds.--BENKING.

Concepts of species.--C. N. Slobodchikoff(Ed.). 1976.Vol. 3, Benchmarkpapers in systematicand evolutionarybiology. Stroudsburg, Pennsylvania, Dowden, Hutchinson and Ross,Inc. 368 pp., illus. $25.00;Les probl•mes de l'esp•ce dans le r•gne animal.--C. Bocquet,J. G•nermont,and M. La- motte(Eds.). Vol. 1, 1976.M•moire no.38, Paris, Soci•tkZool. France.407 pp., illus. 100 Frenchfrancs (about$20.00). (Can be obtainedfrom Soci•t• Zoologiquede France, 195 rue SaintJacques, 75005 Paris, France.).--"Few biologicalproblems have remainedas consistentlychallenging through the past two centuriesas the speciesproblem." This sentenceby Ernst Mayr opensup the volumeon "The Species Problem"published in 1957from a symposiumthe American Association for theAdvancement of Science held in 1955.This book, whichincludes some very importantchapters (those of Mayr, Grant, and Sonnebornin particular)on thespecies concept in differenttaxonomic groups or viewedthrough the lenses ofvarious biological specialists, can surely be considered a landmark publication on this topic. Since 1957, of course,the "species problem" has been dealt with in numerousjournal articles and essays, in severalbooks on taxonomictheory and practice,and in Mayifs monumental"Animal speciesand evolution"(1963, abridgedversion 1970). This 20-yearperiod has witnessed suggestions to use definitions of speciesdiffering in variousdegrees from the biologicalone synthesized by Mayr, and viewschallenging the notionthat speciesare fundamental evolutionary units. More generally, we havelived through a periodof change in the philosophicalappreciation of species,speciation, evolution, systematics, and taxonomy.Yet the "species problem"is apparentlynot solved,for Sokalrecently started an essayon this subjectwith the sentence: "Speciespresent a perennialproblem for biologicalsystematics and evolutionarytheory" (1974, Syst. Zool. 22: 360). The simultaneouspublication of thesetwo booksmight seemto confirmthis view. It mustbe statedat the onsetthat thesebooks are utterlydifferent in spirit, scope,and goal:they simply cannotbe compared.The first, editedby Slobodcbikoff,is a collectionof 21 papersor chaptersreprinted 802 Reviews [Auk, Vol. 94

from the original texts. These are gathered under five major headings, each introduced by a few pages written by Slobodchikoffhimself. The second,edited by Bocquet, G•nermont, and Lamotte, is the first of a three-volume treatise on the problems of speciesin the animal kingdom. It contains eight original essay- chapters on the speciesconcept in various animal groups. Thus Slobodchikoff'sbook is a compilation, whereasthe Bocquet-G(•nermont-Lamottevolume is a descendantof the book edited by Mayr in 1957. Each will be reviewed accordingly, and separately, below. "Conceptsof species"is volume3 in the Benchmarkseries in systematicand evolutionarybiology. (The previous two were both devoted to "Multivariate Statistical Methods.") Slobodchikoff makes his intentions crystal clear in his preface(p. ix): "The selectionsin this book were chosenprimarily becausethey representa point of view or illustrate the historical and philosophicaldevelopment of speciesconcepts." And further: "In making selections,I had two objectivesin mind. One objective was to bring together enough material about the historical background, the philosophicalbases, and the presentviews on speciesto permit critical evaluation of present speciesconcepts within the framework of current biological theory. The other objective was to choosepapers that could be usedby studentsof systematicsand evolutionarybiology to expandtheir understandingof the problemsposed by the speciesquestion." To conformwith this plan, the five parts of the book are: (I) Historical and philosophicalorigins of speciesconcepts; (II) Are speciesreal?; (III) Microbial, parthenogenetic,and paleontologicalspecies; (IV) Speciesin higher plantsand animals;and (V) Alternative concepts. The objectivesseem good, and the groupingsappear logical. But a thorough examination of the chapters or paperschosen for inclusionshows that Slobodchikoff'sobjectives are simply not fulfilled. The "historical and philosophicaldevelopment of speciesconcepts" is offered through the distorting glassof generalessays, so that every historicalpoint of view is, at best, presentedonly second-hand.Furthermore, only English- speaking writers were chosen, so that the reader cannot study directly the schoolsof thought on species concepts dominated by other cultures and languages, especially if these conceptsare recent. Thus the unwary graduate studentwill be led to believethat only English-speakingauthors write about the species today, and will not get an inkling of the thoughtsof German, French, Russian, Scandinavian,or other authors.Finally, if I considermyself a "studentof systematicsand evolutionarybiology," how canI "expand my understanding"of the speciesproblem when noneof the essaysreprinted in Slobodchikoff'sbook is new to me, and when I have in my personallibrary 10 of the 21 chapterseither becauseI own the books(Mayr, Grant, Crowson), or I subscribeto the journals (Science,Systematic Zoology, American Naturalist), or elseI have reprintsof theseworks? For thesereasons, and alsobecause it is too expensive($25.00), I would neither assignthis book to my studentsif I were to teach a graduate courseon the speciesproblem nor buy the book for my own use. "Lesprobl•mes de l'esp•cedans le r•gne animal" owesits origin to the centennialof the Soci(•t(•Zoologique de France in 1976. In view of the importanceof the speciesas an evolutionaryand systematicconcept and unit, and of the lack of a generalbook on this topicin French sincethe publicationof Cu6not'sseminal little book of 1936 ("L'Esp•ce," Doin, Paris, 310 pp.), the Society sponsoreda three-volume treatise on this subject. Volume I appearedin 1976, Volume II is in press,and Volume III in preparation. Chapter I of VolumeI is an "Introductionto the conceptof speciesin the animal kingdom"by Bocquet,G•nermont, and Lamotte. Chapters II-IX are monographson the speciesconcept and problemsof speciesin birds (II), teleost fishes(III), Lepidoptera(IV), Drosophilidae(V), mosquitoes(VI), marine crustaceans(VII), marine bivalves (VIII), and Protozoa(IX). It is a pity that no Englishsummaries are included.Volume II will includeanother seriesof taxon-orientedchapters (including primates, rodents, gastropods,collembola, amphibians, and lizards). Volume III will contain syntheticchapters on various criteria (morphology,biometry, caryology, biochemistry, ethology),on numerical taxonomy, on the speciesconcept in paleontology,on nomenclatural problems,and a generaloverview. Thus whencompleted, "Les probl•mes de l'esp•cedans le r•gne animal" will be a much enlarged and worthy successorto Mayr's 1957 compendium. It would be premature to review this book on the basisof volume I alone, as the entire three-volume text will constitutea unit. Furthermore, asI authoredthe chapteron "The speciesconcept in ornithology,"I will say nothing about it here, but I would like to single out the chapterson Lepidoptera (Guillaumin and Descimon),on mosquitos(Grjebine, Coz, Eluard, Mouchet, and Rageau),and on Protozoa(G•nermont) as outstanding.I would like alsoto statethat the biologistswho wrote chaptersfor volume I, unlike many of their colleagues,do not appearto considerthe "speciesproblem" as a Gordian knot in systematics,seem less concernedby problemsof the definitionof species,and are lessannoyed by the variability in "operationality" of the biologicalspecies concept. On the other hand, they seemto acceptthe multiformity of kindsof species and to live not too unhappily with the difficulty of mergingpure morphologywith reproductiveisolation. In a word, they seemquite pragmaticabout the wholebusiness of the problemof species,without for an instant October1977] Reviews 803 beingless thoughtful or lessphilosophical of its implications.Much asDobzhansky did in hisessay on species ofDrosophila (Science 1972), these biologists find "newexcitement in an old field." I hope"Les probl•mes de l'esp•ce dans le r•gne animal" will receive all the attention it deservesfor it containsmuch food for thought.--FRANqOISVUILLEUMIER.

Atlas des Oiseaux Nicheurs de France.--Laurent Yeatman. 1976. Minist•re de la Qualitg de la Vie-Environnement,Paris, Soci•t60rnithologiquede France. 282 pp., maps, figs., tables. 70 French francs(obtainable from Soci•t• Ornithologiquede France, 55 rue de Buffon, 75005 Paris, France).-- Mapping the distributionof bird speciesis one of the first important stepstowards delineating their geographicalranges accurately. The boundariesof theirranges can thus be delimited precisely, and the total area of distribution estimated. Furthermore, one can make statementsabout the continuity or discontinuity ("patchiness")in distributionand seekreasons for the observedpattern. One can alsosuperimpose maps of thedistribution of severalspecies and estimatethe densityof species(in a fauna, or a groupof relatedspecies, or a "guild")in relationshipto topographyor vegetation.In short,good maps are a preciousand indispens- able tool for biogeographicstudies and generalizations. But what is a goodmap? Probably the mostimportant criteria for "goodness"are: (1) the entirerange of the speciesshould be included, or at least a substantialportion of it, (2) the distributionaldata shouldbe basedon samplingthroughout the rangeor that portionof it beingstudied, and (3) the distributionaldata shouldbe gatheredover a very shortperiod of time (a few years).The first criterionwill insurethat the map includesall or someof the actual or potential geographicalvariation in the ecologyof the speciesbeing mapped.Secondly, the samplingguarantees that the coverageof localitiesis uniformand not arbitrary. And thirdly, the timelimit minimizesthe risk of preparinga compositemap including old and recentrecords, and thus permitsone to gaugethe possibilitythat unusualrecords taken at differenttimes are due to range expansion or contraction during that time span. Many publishedmaps suffer from one or more of several defects. They are usually basedonly on the locality recordsavailable to the worker (either museumskins or valid sightrecords), rarely on sampling. Distributiongaps are thereforedifficult to evaluate:are they real or an artifact of lack of sampling?Such mapsusually show extrapolated boundaries of distribution,which are all tooeasily •:aken as fact; this sort of datum is especiallyimportant in studiesof allopatric or adjacent but not overlappingdistributions. Of coursethe problemis how to obtaina sufficientlylarge number of localityrecords to satisfythe requirements of samplingin a minimum period of time. Clearly only a very large team or group of workers can hope to accomplishthis task. This is an area of ornithology where amateurs and professionalscan fruitfully collaborate, as the present Atlas testifies. In 1969¾eatman, then presidentof the Council of the Soci•t60rnithologiquede France, was inspiredby the example of the British Trust for Ornithology, which was engagedin the preparation of an Atlas of BreedingBirds in Britain and Ireland. ¾eatmandecided that a similareffort could be carriedout in France. From 1970to 1975amateur ornithologists censused bird distributionin France,following a setof guidelines providedby ¾eatmanand hisimmediate collaborators. For eachof 1,100rectangles of about20 x 27 km these observerswere to note on file cards whether each bird specieswas possibly, probably, or certainly breeding.The rectanglesize was dictated by theavailability of the mapsof France,scale 1:50,000, published by the Institut GgographiqueNational. AppendixIV, pp. 276-277, showsthe basemaps and rectangles. From 1970to 1975a total of 1,092 of the 1,100rectangles were actually coveredin the field: only 8 (0.73%) remained totally unstudied. Thus coverageis virtually complete. Progressin coveragefrom 1970 to 1975 is chartedin Fig. 1 (p. X) and on two maps.The oneon p. 2 showsthe coveragein 1970to 1972, and reveals many gaps;the one on p. 3 showsthe entire coveragefrom 1970 to 1975. The gapsare indeed few, but Yeatman admits (p. IX) that "not all maps [i.e. rectangles]have been equally deeply prospected";he remarksthat Sarthe, Dordogne,and Lorraine were the least well studiedregions of France. Thesegaps, althoughreal, canbe clearlyseen on the map p. 3, which thus servesas a standardof coverageagainst which one must compare the maps of bird distributions. In view of the small number of contributors, and of the vast region that was surveyed in 6 years (over 500,000square km), I think that the gapscan be considered minor, and that thedistribution maps published in thisAtlas canbe usedeffectively to obtainan accuratepicture of the distributionof breedingspecies in this part of WesternEurope in the period of study. This feat is considerable.In the United Kingdom, where amateur ornithologistsabound, such a collaborativeeffort can be accomplishedon a very differentscale. Between 10,000 and 15,000 personscontributed to the B.T.O. Atlas, and about 1,500 of them were major contributors(all 3,862 quadratswere covered).¾eatman achievedbetter than 99% coverageof 1,100 rectangleswith only about 500 collaborators.Such an effort would have been unthinkable in France a 804 Reviews [Auk, Vol. 94

decadeago, because the hobbyof bird-watchingwas then virtually unknown.The presentAtlas thus attests to the recentdevelopment of field ornithologyin France,and to the remarkabletalents of Yeatmanand his colleaguesin the Soci6t60rnithologiquede Francefor havingbeen able to enlistso efficiently the helpof amateurs, in a country well known for the individualism of her inhabitants. The bulk of thebook (pp. 4-266)is devotedto theactual 264 distribution maps and an accompanyingtext. The formatis the samethroughout, and the Atlas is thusvery easyto use.The textspecifies the geographical and ecologicalaspects of distribution,indicates fluctuations in populationsize or range,and comparesthe statusof eachspecies in the period 1970-1975and prior to 1936, when N. Mayaud publishedhis "Inventaire des oiseauxde France." The maps show, by one of three symbols(dots of varying diameter)on each rectanglewhether the specieswas recorded in it and if so,whether its breedingwas deemed to be possible, probable,or certain(the criteriafor determinationof thesecategories are givenon p. VIII). The numberof maps(rectangles) where the specieswas notedand the percentagesof eachstatus category are alsogiven. ThusPodiceps cristatu• (p. 5) was recordedfrom 293 maps(26%), and of these,was reportedas possibly breedingin 16%, probablyso in 6%, and certainlyin 78%. Perusingthese 264 maps one seesa number of qualitative trends in distribution.Many of them can obviouslybe correlatedwith someof France'svaried topography, climate, and vegetation, including a long and ecologicallyvaried coastline,several large rivers, estuaries,wetlands, high mountains,lowlands, Mediterraneanmaquis, a largepeninsula (Brittany), a largeisland (Corsica), etc. These maps therefore give usa detailedpicture of arian distributionin a largechunk of WesternEurope. Careful study of themwill not onlybe qualitative, however. As eachmap indicates the 1,100rectangles, it is possibleto makequantitative analysesof speciesdiversity such as congruenceof faunallimits, continent-continent,and continent-island comparisons.These analyseswill lead to many questionsthat can be studiedfurther in the field. Keepingin mind the limitationsof coveragecarefully mentioned by Yeatman himselfin the introduction, andcited in this review,and the inevitablepossibility that errorshave crept in eitherduring fieldwork or duringthe collectionof datato preparethe maps,it is neverthelessclear that themaps in thisAtlas fulfill to a considerabledegree the three criteria for "goodness"set forth earlier. The potentialdrawbacks of having usedfor Francerather large rectangles of about20 x 27 km, insteadof smallquadrats 10 x 10 km asin the Britishand Irish Atlas, are offsetin the publishedpresentation. Whereas each reader can, if he sowishes, carry out numericalanalyses of avian distributionin France, on the solebasis of his copy of the Atlas des Oiseaux Nicheurs de France, this is virtually impossiblefor the U.K. without consultingthe archivesof the B.T.O., for the quadratsare not includedon the British and Irish maps,I supposebecause the scaleof the mapsmakes it impracticalto includethe quadratswithout making the mapsalmost illegible. In my view the publicationof the French and of the U.K. Atlasesmark the beginningof a new era in mappingarian distributionon a largescale. A EuropeanOrnithological Atlas Committeewas founded in 1971,and it is hoped that a EuropeanAtlas will eventuallybe published.In the meantimeYeatman and the Soci•t• Ornithologiquede France are to becongratulated on a finepiece of work,which was subsidized financially, it is gratifyingto note, by the FrenchMinistry of the Quality of Life and of the Environment.--FV•N•OIS VUILLEUMIER.

Bird Taxidemy.--James M. Harrison. 1976.North Pomfret, Vermont, David and Charles.xvii + 67 pp. $6.95.--A small volume on the art of taxidermy designedfor the lay person.The instructionsare accurate and completewith specialattention given to problemscaused by decomposition,slippage, and dirty skins. The illustrations are clear. The only possibleproblem for American readers may be some differencesin terminology,especially for the materials,as this book is written for a British audience.- WALTER J. BocK.

William Beebe. An annotated bibliography.--Tim M. Berra. 1977. Hamden, Connecticut,Ar- chon Book (ShoeString Press).157 pp. $15.00.--This volume, as the title indicates,is an annotated bibliographyof Beebe'swritings from his first article written when only 18 yearsold on the Brown Creeper published in 1895 to his last annual report of the Department of Tropical Researchof the New York ZoologicalSociety in 1962shortly before his death at the age of 84. A biographicalsketch and several photographsof Beebeare included.This is a very usefulvolume for anyoneinterested in the historyof scienceand in the life of thisoutstanding ornithologist, zoo man, explorer,writer andnaturalist.--WALTER J. BocK.