102 PROCEEDINGS OF THE NATIONAL MUSEUM

occurrence of which, ⁱⁿ other regions, is

shown

on the table with the exception thattwospecies areomitted,Aequoreaaequorea, because our knowledge of its several varieties is still so chaotic thatit isnot advisableto

make

anyuseof

them

asyetin.discussions of zoogeography,and the" species-incerto" of

Murbach

and Shearer (1903), a peculiar codonid, because

known

from a single poorly preserved specimen.

Distribution ofBeringSeaspecies.

NO. 1946. PACIFIC

MEDUSAE AND SIPHONOPHORAE—BIGELOW. 103

of the Aleutians, and the Sea of Okhotsk alone, that is, Meator rubatra and CalycopsisnematopJiora, bothdescribed herefor thefirst time.

And

thoughthesecond

may

fairlybeassumedtoberestricted to theBerii:^Searegion, becauseitis a surfacespecies and has not been taken elsewhere although the genus to which it belongs is

widelydistributed,thefirstiscertainlya

member

ofthemesoplankton, forwhich reason itis unsafe to assumethat itwill notprove to be

much more

widely distributed than

now

appears,

when

thebathy- metric province towhich it belongs is betterknown. "The

same

is true of

Pandea

rubra, recorded from Bering Sea and from off the coastof British Columbia,and it is questionablewhether the single

Scyphomedusae

peculiar tothe BeringSearegion, Haliclystus stejnegeri Kishinouye,isreallyseparablefrom H.auricula.

The

remainingspeciesare easilyclassifiedas to their relationships, systematic andgeographic. Inthefirstplace, therearefour"inter- mediate"cosmopolitan species, belonging that is, to themesoplank- ton,Halicreaspapillosum,Aeginuragrimaldii,PeriphyUahyacintJiina, Atolla wyvillei, and one cosmopolitan surface form, Aurelia aurita, while two others, Solmissusincisa and Aegina rosea, ifnot cosmo- politan, are at leastwidely distributed both in the Atlantic and in theIndo-Pacific.

Then

one, BougainviUea hougainviUei,is probably identicalwiththeother

member

^of the

same

genusmentionedinthe

list, B. superciliaris. Sarsiaprinceps, S. eximia, Hyhocodonprolifer,

BougainviUeasuperciliaris,RathkeahlumenhacJiii, Catablemavesicaria, Tiaropsis diademata, Eutonina indicans, StauropJiora mertensii, PtycTiogenalactea,Aglanihadigitale,PantacJiogon JiaecTceli,Botrynema eUinorae,Aeginopsis laurentii, Haliclystus, and Cyanea capillata var.

capiUata, are characteristic

members

of the

medusa

fauna of the boreal Atlantic, or of the Arctic Ocean; Aurelia limhata probably occurs in Greenland waters (Vanhoffen, 1902a) and Sarsia japoTi- ica

may

beidenticalwith an Arctic species,S.Jiammea, while Cata- Uerna multicirrataisa

member

of a genuswhich,as heredefined,is

known

onlyfrom the boreal Atlantic and from the Arctic Ocean.

Six species, on theother hand, clearly have an Indo-Pacific origin.

These are Heterotiara anonyma, which ^is widely distributed in the intermediatedepths of the Indo-Pacific, butisnot know^nfrom the Atlantic; ProhoscydactylaJlavicirrata, a Pacific species which finds itsclosest ally intheonly other

member

^ofthe genus,ornata,^oneor other variety of which is recorded from various locaHties in the tropical Atlantic and the Pacific and Indian Oceans; Gonionemus vertens var. depressum, which is represented in the north Atlantic

by

a close ally; and CTirysaora lielvola, which is so far

known

only fromthetwo sides ofthenorthwesternPacific,thoughit iscertainly

11haverecentlyexaminedtwoexcellentspecimenso( flavicirratafrom Puget Sound,findingthat the modeofbranchingofthe canalagreesvery wellwith Brandt's(1838) figure; thisseparatesthem,as Isup- posed(19096),fromornata.

104 PROCEEDINGS OP THE NATIONAL MUSEUM.

vol.44.

veryclose to theAtlantic G. hysoscela,ifindeeditisseparablefrom

it at all. 0. melanaster, too, is so far

known

onlyfrom the north- westernPacific;andthePelagiaisa representativeofatropicaland subtropical genus, a casualvisitorfromthesouth. _,

Thus, as might have been expected on oceanographic grounds, it is evident that the

medusa

faunaof theBeringSearegionhas been recruited from two directions, there being a constituent from the

warmer

waters ^of the Pacific on the one

hand

and, on theother, a

more

importantone,numerically, fromthecoldArcticwaters,which

must

haveenteredBering Sea

by way

^ofBeringStraits.

Most of the leptoline species of northern origin are

known

from the region extendingfromCape

Cod

tosouthernLabrador, whileone, Eutonina socialis, occurs in theNorth Sea; thatis, theyareboreal rather than purely Arctic, though several are

known

from Arctic stations; and this is

what we

might have expected, for in

summer

thetemperatureofthe surfacewaters ofthesouthern partofBering Sea is from 50° to57°; thatis,aboutas

warm

as thewatersof the northern parts of theGulfof Maine on the

New

England coastand considerably

warmer

thanthe surfacewatersoftheLabradorcurrent off Labrador.

And

for tliis

same

reason itis probable thatcollec- tions

made

during the late

summer

would contain a greater

number

of southern ^species, for the Fisheries steamerAlbatross records are limitedto

May

andtheearlypartofJune.

The

Bering Sea list includes four siphonophores, all long-known species.

One

of them, DipJiyes arctica, previously recorded only from Arctic and boreal regions (19il&),was taken regularlyon the presentcruise in coldwaters, but was entirely absent in the

warm

waters ^of theEastern Sea and of

Kuro

shiro. Like the Arctic and borealmedusae, itprobablyreached theBering Sea regionfrom the north,i. e,

by way

ofBeringStraits. But, asDoctorMoser,

who

is

now

workingonthe siphonophoresofthe

German

SouthPolarExpe- dition, writes

me

that theGauss took itin

many

localities, further discussion of its distribution is best postponed until her data is

published.

The

three othersiphonophores are all species ofwide distribution in bothAtlantic and Indo-Paoific, Rosaceaplicata

and

Biphyestrun- catabeing

known

from tropical aswell asfrom temperate stations and Vogtiapentacantliafrom the

Bay

ofBiscayand from theMedi- terranean.

But

as noneof

them

has everbeenrecorded from Arc- tictemperatures, itissafe toassumethatthey colonized BeringSea from the south.

They

were all taken there so regularly that they are tobe consideredcharacteristic

members

ofits pelagic fauna.

The

collection suggests the probability that certam species of

medusae

belongingtointermediate depths

may

beconfinedtoArctic- boreal regions; a generalizationofgreat zoogeographic interestifit

NO. 1946. PACIFIC

MEDUSAE AND SIPHONOPHORAE—BIOELOW.

105 proves to be wellfounded. This tentative conclusion (ofcourseit

may

have tobemodified) isbasedontheoccurrenceofPantachogon Tmeckeliand of Crossotabrunneavar. norvegica. Ifthesespecies

had

beenrepresented

by

onlyafewscattered captures,

we

mightwell hesi- tate touse

them

asinstancesofgeographic range; but both weretaken inconsiderablenumbersat 10 stations each,and^theregularity of their distributionisstriking

when we

observe that only 14 hauls with the intermediate netfrom300fathoms were

made

southofthe

Shumagin

Islands, in BeringSea and in the Sea of Okhotsk,in 11 of which oneorotherof thetwospecies

was

taken (insixhauls thetwo were taken together; each

was

likewise taken in the trawl).

On

the other hand, both were conspicuously absent in the intermediate hauls

made

in Japanese waters, nor are they represented in the extensive collections

made by

the Albatross off the coasts of Cali- fornia, British Columbia,

and

southern Alaska, which have passed through

my

hands. In the Atlantic, too, these two species have been taken only at far northern stations, though the expeditions of the National, of the Valdivia, and of the Prince of

Monaco

might have been expected to reveal

them

in the

warmer

parts of the Atlanticwere they as

common

there as inBering Sea,oras are such generaas Halicreas, PeripJiylla,andAtolla.

And

thefact that P.TmecJceliisrepresented

by

adistinct,thoughallied,species,scotti,in the AntarcticandC.brunneavar.norvegica

by

a recognizable varietyin tropical regions,isalsostrong evidence that thetwoarerestricted to northern regions.

The

collection

made by

the Michael Sars,

now

being studied

by

DoctorBroch,

may

throwfurtherlightonthissub- ject, but

we

are probablysafe in assuming thatboth Pantachogon and Crossotaentered BeringSea fromthe north.

Botrynema ellinorae is likewise

known

only from "intermediate"

hauls inBering Seaandin the ArcticOcean; but the recordsofits captures aretoofewtowarranttheassumptionthatitisconfinedto highlatitudes,while thefact thatits closeally, brucei, isAntarctic, suggests that the genus, atleast, wiU be foundin the deeperwater layers of low latitudes. It is not improbable that the two

new

"intermediate" species,

Pandea

rubra and Meator rubatra,

may

belongtothe

same

category,for theiroccurrenceinthenorthwestern Pacificclosely parallels thatof Pantachogon

and

Crossota; but, as pointed out above, itis best to suspend judgmentonthispoint for the present.

We

can not trace the cold water

medusae

southward along the west and east sides of the northwestern Pacific aswellas

we

could wish, because

we know

verylittle aboutthe

medusae

oftheAmeri- cancoastbetween the Aleutiansand Puget Sound.

But

thecollec- tionsfromthe latter region, recorded

by

A. Agassiz (1865) and

by

Murbach

and Shearer (1903), and aseries in the

Museum

of

Com-