Contributions to Zoology, 72 (I) 39-72 (2003)

SPB Academic Publishing bv, The Hague

Ingolfiellidea (Crustacea, , ): a phylogenetic and biogeographic analysis

Ronald Vonk & Frederick+R. Schram

Institute for Biodiversity and Ecosystem Dynamics, University ofAmsterdam, Mauritskade 57, 1092 AD

Amsterdam, The Netherlands

Keywords:: , phylogeny, biogeography, new genera, paleogeographic maps, bibliography

Abstract Paraleleupia

Ingolfiella

Additional descriptions 63 The suborder consists of 39 named Ingolfiellidea currently spe- Biogeographic analysis 65 cies. An historical overview is presented and phylogenetic and Discussion 67 biogeographic analyses are made. The result ofthe phylogenetic Acknowledgements 71 analysis the definition of new within an suggests two genera References 71 African and freshwater group, namely Paraleleupia n. gen.

Proleleupia n. gen. Re-examinationof a supposedly Italian re- lict species, Metaingolfiella mirabitis, with the aid of SEM Introduction techniques reveals a half-fusion ofthe head region with the first pereionite.The issue of the function of the ‘eyelobe’ is addressed and an explanation presented after examining with SEM such The ingolftellidean amphipods are not abundant in lobes in different additional species. Furthermore, descriptions regards to the numberof species. To date, some 39 are based given onthe type-material ofMetaingolfiella mirabilis, species are recognized, a remarkably low number Trogloleleupia eggerti, Trogloleleupia leleupi, lit- the wide-ranging ecological conditions toralis, considering I. tabularis, I. margaritae, I. quadridentata, and I. abyssi. An in which they occur. It is also a low number when attemptto relate the geographic distribution and ecological to other of with characteristics to the phylogeny of the Ingolfiellidea results in compared family groups amphipods a habitat paleogeographic scenario that points to a freshwater sub- partly overlapping requirements such as terranean for origin the group that dates from at least Triassic bogidiellids, with 110 species (Koenemann & Hol- times. and singer, 1999) crangonyctids (250 species, pers.

comm. Koenemann).

The first specimens ofIngolfiellidea were reported Contents Hansen 1903. He classified them by in in a new

suborder, a rank they have continued to hold in Introduction 39 most publications ( Bousfield & Shih, 1994; Ruffo Methods and material 42 & 1989; Martin & Davis, Character Vigna Taglianti, 2001) descriptions 42 several the Cladistic despite objections over years (Dahl, 1977; analysis 57

Results Bowman & Barnard & ' 58 Abele, 1982; Karaman,

Systematics It is to this small 60 1983). a challenge compare group Taxonomic diagnoses derived from cladistic analysis 60 of other well-defined to larger groups Ingolfiellidea 60 find and try to comparable habitat preferences and Metaingolfiellidae 61 overlapping biogeographic distribution patterns. 61 From consideration of the literature a picture Stygobarnardia 61

Trogloleleupia 61 emerges of the uniqueness of ingolfiellideans within

the entire Proleleupia 61 world with regard to their

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 40 R. Vonk & F.R. Schram -A phylogenetic and biogeographic analysis

I. World of distribution of Black dots = marine localities. = brackish conditions. Fig. map Ingolfiellidea. Open stars water Open

circles = fresh water localities.

incredibly diverse habitat tolerances. No other small Congo, later ranked in a separate : Troglole-

taxon of crustaceans is found in the soft mud of leupia leleupi (Ruffo, 1951). Compared to the 2.5

well the deep-sea floor, as as in high mountain fresh- mm specimens for previous taxa this African cave

in stood 14.5 maximum water river beds, or subterranean fresh, brackish lake species out at mm length. and marine interstitial waters of continental ground Then more species were discovered in Macedonia:

waters and continental shelves (Fig. 1, map) Ingolfiella petkovskii S. Karaman, 1957, and I. mace-

The study ofthe first species of Ingolfiella marked donica S. Karaman, 1959. S. Karaman character-

this habitat diversity. Hansen (1903) described two ized their rarity when he stated that he had searched

species, Ingolfiella abyssi, from 3521 m. deep sea through several thousand samples from ground water

bottom out of a trawl of 1 litre of mud in the Davis and wells. He concluded that Ingolfiella is either

of and I. from coral lives in inaccessible Kara- Strait, east Greenland, littoralis, very rare or biotopes (S.

sands on the shores ofThailand. Several later years man, 1959).

anotherspecies, I. acherontis (S. Karaman, 1933), Meanwhile, another species was found in South

was described from a groundwater well in Skopje, America in a new habitat, the coastal ground water

Macedonia. The material ofI. acherontis has been in a coarse shingle beach. In the description of I.

few lost and only a incomplete drawings remain. ruffoi Slewing, 1958, Slewing posed a dilemma

S. Karaman apparently was not aware of the work particularly relevant to the work underlying this

by Hansen, 1903. Therefore he describedBalcanella publication. He asked whether the large African

acherontis in form the and n. gen. n. sp. a new family Balcanel- cave was ancestral to smaller marine

lidae. freshwater if it the other around. forms, or was way

After the Second World War expeditions focused Another type of habitat was added to the list

more on environments that with the collection of of britannica were difficult to access, 20 specimens I.

and the of majority species ofingolfiellideans were Spooner, I960, from the shell gravel at the sea

collected in half the last of the twentieth century. bottom at 46 meter depth off the south coast of

A in large species was found cave waters from England. Spooner remarked on the true deep sub-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 41

surface occurrence of these ingolfiellids in contrast From the same environment, but a little deeper at

in of with other small malacostracans that more typically 8 m below the surface the Gulf Naples, Italy, inhabit only the first few centimeters of these gravels. Schiecke (1976) described I. ischitana. Stock named

tolerance revealed within Euryhaline was one 6 species from islands off the coast of Venezuela.

I. manni species with the discovery of Noodt, 1961, These species are: I. (Gevgeliella) putealis Stock,

taken from both brackish and fresh ground water 1976, in a slightly brackish well on Bonaire (Dutch in Chile. In addition, I. manni was found not only Antilles); I. (Gevgeliella) fontinalis Stock, 1977,

at sea level, also at 800 m. Later I. but uspallatae from a spring on Bonaire; I. (Gevgeliella) tabularis

1965 was described from in the Noodt, Argentina Stock, 1977, from marine sands situated below a

riverbanks of a at 2000 m in the Andes floodplain cave entrance on Bonaire (also recorded on Cura- mountains. cao); I. (Hansenliella?)quadridentata Stock, 1979,

In at this two freshwater Europe time, species from coarse coral sand of a submarine flat, Curacao; were reported from the interstices of river alluvia I. (Trianguliella?) grandispina Stock, 1979, pumped of southern France; I. catalanensis Comeau, 1963 in from brackish ground water gravel at a cave and I. thibaudi Coineau, 1968. This marked the entrance; I. (Gevgeliella) margaritae Stock, 1979, of where subsur- discovery more species special for from a capped freshwater well used drinking face habitats were sampled in a conscious effort to water on Isla de Margarita (Venezuela). These south- A 2.3 uncover new taxa. large cm ingolfiellid was ern Caribbean species were sampled either by div-

found in a well in first described as Leleu- Namibia, in shallow coastal from ing water, or pumped up piella eggerti, Ruffo, 1964, later renamed by Ruffo beach interstitia and oligohaline water reservoirs (1974a) as Trogloleleupia eggerti. more inland. In addition, occurred in wells From the Indian Ocean I. xarifae Ruffo, 1966 and small cave chambers with brackish water or and I. & kapuri Coineau Rao, 1973 were reported water with frequently and strongly fluctuating sa- from the Maldives in shallow coral sands, and from linities. the Andaman and Nicobar Islands in intertidal mud- Stock utilized subgeneric names, but the split- shell dy debris, respectively. ting of the genus Ingolfiella into smaller units had A second deep-sea species, I. atlantisi Mills, 1967, already begun with that earlier discovery of In- was collected 4700 at more than m depth in the golfiella (Balcanella) acherontis Karaman, 1933. North American Basin of the Atlantic Ocean. This Nevertheless, Karaman’s original effort did not I. species probably occupies the same habitat as useful due to subsequent workers’ lack abyssi prove very and was found amidst typical abyssal fauna of in male or female specimens cases where both elements, presumably residing “...in the flocculent sexes were needed to make critical distinctions. and relatively oxidized upper centimeters or two Consequently, new species were difficult to incor- on the surface of the deep-sea ooze.” (Mills, 1967). into the existing system. Back in porate Europe, an environment equally diffi- Ruffo (1985) described another large ingolfiellid cult to access as the deep sea was tapped, namely the from Namibia’s freshwater reservoirs, Stygobar- ‘fossil water’ of in a 50 m deep well Italy. Here The the nardia caprellinoides. most striking aspect of aberrant Metaingolfiella mirabilis Ruffo, 1969 the this was resemblance of the form of the was described, for which a new family was neces- to that of the Italian In sary, the cephalon Metaingolfiella. Metaingolfiellidae. Later efforts to recap- of ture this Metaingolfiella the fusion the first segment (pe- rare species were never successful (pers. reionite) with the was described as com. Not and cephalon fully Ruffo). surprisingly, the area south have discovered east of complete (although we after SEM Yugoslavia has yielded more specimens of that this be the I. petkovskii investigation appears not to case), and the new I. vandeli species . Bou,

'970. while in the head form with a were Stygobarnardia typi- They reported from wells and river

sediments cally small cephalon could be interpreted as in a on the Greek mainland and the large islands partial stage of the fusion. of the Peloponnesus and Euboea. Another One of the Islands the coast discovery occurred in the littoral sands larger Canary facing of Table of West Africa, Fuerteventura, yielded Ingolfiella Bay, Cape Town; I. berrisfordi Ruffo, 1974.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 42 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

work the has concentrated similis Ronde-Broekhuizen & Stock, 1987, from ous on Ingolfiellidea on

freshwater To this situation an oligohaline well. Ingolfiella fuscina alpha-. remedy we un-

Dojiri & Sieg, 1987, was found in the Gulfof Mexico dertook the following:

of andoff the coast of South Carolina in box core and a. Performed a cladistic analysis phylogenetic

- 151 based grab samples ranging in depth from 17 m. relationships within the Ingolfiellidea on

This the character matrix. was first time a species from the bottom a comprehensive ofthe On Ber- b. Restudied material of mi- sea was reported over a large area. type Metaingolfiella

T. muda, I. longipes Stock, Sket & lliffe, 1987 was rabilis, Trogloleleupia eggerti, leleupi, Ingol-

I. I. I. collected in a brackish water cave pool. More large fiella abyssi, littoralis, tabularis, qua-

from and I. de- ingolfiellids came Namibia: Trogloleleupia dridentata, margaritae to add more dracospiritus Griffiths, 1989 and T. gobabis Grif- tails to former descriptions.

km the lobes of fiths, 1989, were found in cave lakes, 350 apart c. Compared cephalic Ingolfiella pu- from each other. tealis and I. ischitana, using SEM photography,

anatomical and func- The Southwest Pacific yielded no ingolfiellids to lend insight into the until Lowry & Poore (1989) described I. australiana tional role this structure plays. andI. bassiana from the Bass Strait at, respectively, d. Re-examined the type material of Metaingolfiella

another 85 m depth in sandy shell, and 121 m at mirabilis to determine the degree of fusion

the locality in the strait. A cave on a small offshore between head and the first pereionite. with the island of Sardinia, Italy, brought to light I. cottarellii e. Performed a biogeographic analysis

Ruffo & Vigna-Taglianti, 1989 from a freshwater results of the cladistic analysis and explored

and of pool. The most recent of the large, cave dwelling, palaeobiogeographic patterns the timing

African ingolfielids is Trogloleleupia nudicarpus evolutionary events in the deep history of in-

Griffiths, 1991 from subterranean waters in west- golfiellideans.

in ern Namibia. It was found crawling on stones

60 shallow water in a small pool m below ground surface. Methods and material

I. canariensis Vonk & Sanchez, 1991, was collected Bou-Rouch with a biophreatic pump on We employed 43 characters to analyze patterns of several beaches along the north coast of Tenerife, relationships throughout the Ingolfiellidea. Some

Canary Islands, and in bottom debris of an anchialine 30 characters are multistate, while 13 features are cave on El Hierro, another island in the geologi- binary. Following here is a list of the characters

of cally younger part of the Canary group. From the employed and an explanation the several alter-

of km the I. in island Madeira, some 700 to north, native states we have used constructing a matrix

in the unguiculata Stock, 1992 occurs coarse sand (Table I). Given the ‘reduced’, worm-like body plan bottom of In I. 43 an anchialine lagoon. Slovenia, of these animals our characters essentially cov- beatricis Ruffo & Vonk, 2001 was described from ered all aspects of the recognizable anatomy, so a single specimen caught in a feeble flowing cave there has been no emphasis of one aspect of mor- stream with a hand net. phology over another. The characters we used are

Of course, more species will be discovered, and as follows. we must wait to see how far the ecological toler- ance and of this of biogeographical spread group

be amphipods can stretched. Already there are re- Character descriptions ports of discoveries in fresh- andbrackish waterof

(Shokita, sediments Japan 1992) and in marine of 1. Ocular (cephalic) lobes the William Prince Sound, and the = Alaska, Virgin state 0 developed

Islands, Caribbean (Cadien, that are pers. comm.) state 1 = reduced

to be confirmed. yet state 2 = absent

From the above it is clear that almost all previ- In one of the out-group taxa chosen, Mictocaris

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 43

halope, this character is termed the eyestalk. Its flagellum is a consistent character throughout the

informative towards location is not between antenna 1 and antenna 2 on Ingolfiellidae and as such not

the front in the between margin of the cephalon, as Ingol- distinguishing species or species groups

the of the fla- fiellidea, but is rather located a little higher, flank- within the family. However, length

the Still consider whole does in few and ing peduncle of antenna 1. we gellum as a vary a cases,

this possible remnant of a stalk in Mictocaris as an the differenceswere scored. Out-groups have more

homologous feature with the typical ingolfiellidean segments, as is common in most peracaridans, and

because antennal cephalic lobe, simply no other function can thus bear a relatively long flagellum (state

be ascribed with certainty to this lobe at the mo- 0). Absolute measurements of the flagellum depend

ment. Lowry and Poore (1989) observed that three on the size of the individual. With a ratio character

peracaridan orders have with like this it is often difficult to divide continuous representatives eye-

stalks or remnants of stalks, i.e., the mysidaceans, distributions into integral states. Still an effort is the spelaeogriphaceans, and the mictaceans. How- made to distinguish between longer and shorter

ever, they conclude that the typical position of the antennae as our goal was to score as much charac-

reduced stalks, which are often in the form of scales ter differentiation as possible.

or pointed lobes, lies at the base of the first an-

tenna and at the rostrum. This is different from that 3. Antenna 1, accessory flagellum

= seen in the Ingolfiellidea, where, as noted, the lobes state 0 four segments

reside = between the first and second antennae. There- state 1 three segments

2 = tore, they do not regard these features as homolo- state two segments

and thus of The is on the inner gous not subordinal importance. accessory flagellum positioned

However, spelaeogriphaceans and mictaceans do side of the flagellum (Fig. 3d, 5k, 7c) and could be

not have the lateral compressed head of ingolfiel- called the “inner” flagellum, if we want to keep lideans (and of most amphipods) and the position the terminology that would be comparable to early, of the lobes could easily have shifted ventrally when fossil finds of diverse crustaceans. The positional

that SEM Mictocaris the would be the case. We have made pho- homology is important here for in

tographs oftwo ingolfiellid species, Ingolfiella ischi- outer flagellum has eight segments and the inner

tana and I. We ask if the (Fig, 2a) putealis (Fig. 3a,b), in which has four segments. might accessory the difference between the developed state of the flagellum in amphipods is comparable to the outer

lobe in the former (Fig. 2, b-e) and the reduced or inner flagellum of Mictocarisl

state is visible. “The biramous is retained in the (Fig. 3, a-c, e, f) in the latter clearly primitive origin In both lobe small that cases, there is a neat fit of this be- form of a accessory flagellum arises from

tween the bases end of the of the first and second antennae, the peduncle,” (Lincoln, 1979, p. 16). In and this suggests that the original function (eye- Mictocaris, the eight-segmented “outer” flagellum stalk) has been replaced by a new function. The bears the aesthetascs, which wouldsuggest the four-

lobes close off the “inner” which lacks aesthe- otherwise open space between segmented flagellum, the antennae to mud particles. Thus the appressed tascs, is equivalent to the accessory flagella in amphi-

lobes and the aesthetascs. rostrum form a tight seal around the pods, also lacking

protruding antennal peduncles, preventing fine gra- nular material from fouling the head region. 4. Mandibular palp

state 0 = present 2 - Antenna I, flagell um state 1 = vestigial

state 0 = longer than basal peduncular segment state 2 = absent

state 1 = medium length, more than half of basal The mandibular palp is lacking in the Ingolfiellidea.

peduncular This that segment implies the palp is not necessary towards state 2 = short, less than half of basal peduncular either securing and moving larger food particles to segment the mouth and the molars, or in cleaning and groom- The low number of segments (4) on the antennal ing, as is typically the case for the larger palps seen

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 44 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

3

1 1 1 1 1 2 2 3 2 3 2 4 2 3 4 4 4 2 2 2 2 2 2 2 2 2 2 2 2 2 2 4 2 2 4 2 4 2 Gnldac

2 0 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 anats

1

1 0 0 0 0 0 0 2 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 lent

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 coxa

9 12 1 1 1 7 1 0 0 1 1 2 2 1 0 1 2 2 1 7 1 1 2 0 1 0 7 1 7 1 7 2 1 7 7 7 7 7 2 1 seam 1

18 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

|

7

0 1 2 2 0 3 4 3 3 3 3 0 4 4 3 4 1 1 3 4 4 1 1 2 2 2 2 3 2 1 1 4 2 2 3 2 7 2 1 MxDdao2thor

6

1 1 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 7 1 1 Mxpfps

5 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 mxDDal

4

0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 0 0 0 0 0 0 7 0 1 Mxolob

3

0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 1 1 Mxoba;

2 3 2 / 1 1 1 1 3 2 1 2 2 / 2 7 2 7 3 3 2 3 ? 7 3 0 1 1 1 1 1 1 2 3 7 2/3 2/3 2/3 2/3 2/3 2/3 1 Mx2 1 1

1 1 2 1 3 2 1 1 1 0 1 0 0 1 0 1 1 1 0 0 2 2 1 1 1 0 1 2 1 1 7 2 1 1 2 2 7 2 1 Mxloa!

0

0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 1 1 0 0 1 1 1 1 2 7 0 0 1 0 1 1 1 0 2 1 7 1 1 Mxloa

I 1

5 4 4 5 5 4 5 9 Mxlso 0 2 1 3 2 4 4 5 4 5 4 4 4 5 4 4 4 4 4 4 4 4 4 4 5 7 4 4 4 4 7

ols 2 2 2 2 2 2

0 2 2 2 2 2 / / 2 / 2 2 2 2 7 2 2 2 2 / / 2 7 8 1 1 / 1 1/2 1/2 1/2 1/2 1 1/2 0/1 1/2 1 1 1 1 1 1 Mxl 1 0/1/2 0/1/2

ol

7 1 0 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 0 1 1 0 0 0 1 7 1 0 1 1 0 1 0 0 0 1 7 1 Mx1

il

0 2 3 3 1 2 2 2 3 4 3 3 4 2 3 4 4 2 7 2 4 4 4 4 2 3 4 4 3 7 3 6 1 5 4 1 1 2/3 1 Mxl

1 1 1 1 1 1 1 1 1 5 molar 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

2 2 2 2 2 2 2 4 Mdoalo 0 0 0 0 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

acc 1 2 1 1 3 0 2 2 2 1 1 1 1 1 1 1 1 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1

A1

fl.

12 1 1 1 1 2 0 0 0 0 0 1 1 1 1 1 7 1 1 1 1 1 1 1 0 1 1 2 1 1 1/2 1 1 1 1 1 1 1 1

1 A1

lo.

0 2 2 1 0 0 2 2 2 1 1 2 2 0 2 2 2 2 2 2 0 0 0 0 0 2 2 1 1 0 0 0 2 2 2 2 2 0 0 eve

soyeri chilensis mirabilis caprellinoides eggerti gobabis nudicarpus halope arista australiana

dracospiritus leleupi berrisfordi catalanensis margaritae petkovskii quadridentata unguiculata uspallatae beatricis Mictocaris Bogidiella Pseudingolfiella Pseudingolfiella Metaingolfiella Stygobarnardia Trogloleleupia Paraleleupia Proleleupia Ingolfiella bassiana britannica canadensis cottarelli fontinalis fuscina grandispina ischitana longipes macedonica manni putealis ruffoi similis tabularis thibaudi vandeli xarifae

I T. T. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. I. 1. 1. 1. 1. 1. 1. 1 1. 1. 1. 1. 1. 1. 1. 1. 1.

0 2 3 4 5 6 7 8 9 1 2 1 1 2 3 4 5 6 7 « 9 11 20 24 25 26 27 28 29 30 32 33 34 35 35 37 35 Table 1 1 1 1 1 1 1 1 1 2 2 ■23 3

1

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 45

multiple

to refer

0 1 0 1 1 1 43 1 0 3 4 3 3 3 3 3 3 1 3 3 3 3 3 3 7 3 3 3 3 3 3 3 1 1 3 3 3 1 3 cell

I fus one 42 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 U3 in

I

1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 4 1 1 1 1 1 1 1 1 7 1 U3 scores

I 1

0 io 7 1 1 1 0 2 2 1 1 2 2 2 2 2 2 1 2 1 2 1 1 1 4 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 U1 1 Multiple 9

0 2 2 2 1 1 1 3 U1-2le 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

1

S

0 1 3 3 3 2 2 2 1 2 2 2 2 2 2 2 2 7 2 2 2 2 2 2 2 2 U1-2si 1 1 1 1 1 1 1 1 1 1 1 2 1 (Altivec). 1

7 ram 0 3 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10 0 0 0 0 0 0 0 0 U1 b

1 4,0 0 36 0 1 Plforml 1 0 2 2 2 3 3 2 3 3 2 2 2 2 2 7 7 2 2 2 7 2 2 2 3 2 3 3 2 2 2 2 2 2 2

I '

5 PAUP 0 0 0 3 PI1-3 0 0 0 0 0 0 0 0 0 0 7 7 0 7 1 0 0 0 0 0 0 0 7 7 0 6 0 0 0 1 0 0 0 0 0 in I

34 0 2 0 0 0 2 0 0 0 0 0 2 2 0 2 2 2 0 0 0 0 1 1 0 1 1 P3-7te 1 1 1 1/2 1 1 1 0 1 0 1 1 I analysis 3

3 1 0 0 0 0 0 0 0 0 0 0 0 P5-7de 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

I

2

0 0 0 0 1 1 3 P3-7dc 0 0 0 0 0 0 0 1 1 1 0 0 1 1 1 1 1 0 0 0 0 0 1 7 0 0 0 1 0 0 1 0 parsimony

1 a

0 0 1 3 1 2 2 2 2 2 2 2 2 2 2 2 P3/4C 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 1 2 1 2 2 in

0

- 3 . 1 1 1 1 1 1 1 1 1 0 0 0 0 1 . 0 1 0 0 0 0 0 0 0 1 1 1 1 used Gn2oa 0/1 0/1 1 1 1 1

0 29 0 2 2 0 0 Gn2siz 2 0 2 0 2 0 0 ? 1 0 2 2 7 1 7 0 7 7 1 7 1 0 7 1 0 7 7 0 7 7 7 0 I characters a

0 2 0 0 0 3 3 0 2 0 2 2 2 3 2 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2 Gn2tio 2 2 1 2 2 2 2 2 the 1

7 2 0 0 ,2 0 0 0 / Gn2da 0 3 1 4 2 1 1 1 1 4 4 1 1 2 3 2 1 3 4 1 2 2 2 3 3 2 2 1 1 1 3 3 J- 1

6 represents 0 2 0 0 0 0 0 0 0 0 0 3 3 3 0 0 2 0 2 3 2 3 3 0 0 3 3 2 2 3 3 0 3 Gn2do 1 2/3 0 1 3 2

I

0 25 0 0 0 3 1 2 0 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Gn2casl 2 0 4 0 0 1 1 0 1 7 0 0 (1-43)

I

4

0 2 0 0 0 2 2 2 3 2 Gn2oasl 3 3 1 0 1 1 1 1 0 3 3 1 3 1 0 3 1 0 1 1 1 1 1 1 1 1 1 1 0 figures

of

row caprellinoides top character. soyeri chilensis mirabilis eggerti gobabis The halope arista nudicarpus australiana one

in dracospiritus berrisfordi canadensis cottarelli margaritae petkovskii quadridentata matrix. Mictocaris Bogidiella Pseudingolfiella Pseudingolfiella Metaingolfiella Stygobarnardia Trogloleleupia Paraleleupia leleupi Proleleupia bassiana britannica catalanensis fontinalis fuscina grandispina ischitana longipes macedonica manni putealis ruffoi similis tabularis thibaudi unguiculata uspallatae vandeli xarifae beatricis I1 Ingolfiella T. T. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. present

t 2 0 2 3 4 5 6 7 8 9 S Table 3 4 6 1 6 7 8 Table 9 11 20 22 23 24 25 26 27 28 29 30 32 33 34 36 37 38 1 1 1 1 1 1 31 3 2 1 1 1 2 Character states

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 46 R. Vonk & F.R. Schram A phylogenetic and biogeographic analysis

in malacostracan crustaceans. Fine food particles state 2 = dentate

in this One the the make up the diet line of reasoning. genus When comparing tips of robust apical setae

in the Ingolfiellidae, Stygobarnardia Ruffo 1985, on the distal margin of the outer lobe of maxilla 1,

rudiment of the mandibular and clear differences possesses a palp is were noticedbetween several spe-

In scored as vestigial (1). cies. most species the inner seta stands apart from

This reduction and/or complete loss of a man- the others and is implanted on the lateral/submar-

dibular occurred in of inner side of it is palp many groups crusta- ginal, the plate. Often clearly

ceans (Richter and Scholtz, 2001) and shows a great dentatewith a little comb, at other times it is simple.

variability within families and generaofAmphipoda.

For mandibular 8. 1 instance, a palp may be present or Maxilla outer lobe, outer setae (Figs. 5a, 6j,

and absent within the Ampithoidae Hadziidae, or m)

in Even 0 = it may vary length as in Metacrangonyx. state simple

so, this phylogenetically uninformative character state 1 = bifid

is retained here to emphasize the special situation state 2 = dentate

and in Stygobarnardia to be available for use with “Simple” setae are smooth, spiniform setae with-

of possible finds new taxa. out hooks. In case more states are present, a multi-

state score is used (0/1 or 0/1/2). Comparison with

Mictocaris 5. Molar process size the out-group halope seems to indicate

0 = state well-developed lobe that an earlier condition might have resembled an

1 = undifferentiated state vestigial peg or spine tuft of soft setae (Bowman & Iliffe,

All Ingolfiellidea share the reduced state (Figs. 5c, 1985)

the 61, n, arrows). In combination with absence of

mandibular palps, this feature perhaps points to a 9. Maxilla 1, outer lobe, setae orientation

lack of need for chew = structures to up large par- state 0 continuous row

ticles of food. state 1=7 + additional inner seta

state 2 = 7

6. Maxilla 1, inner lobe setae (Figs. 5a, 6j, m) state 3 = 6 + additional inner seta

state 0 = numerous state 4 = 6 (Fig. 5a)

state I = four or more state 5 = 5

2 = the state three The setae on distal margin of the outer plate of

state 3 = two the maxilla I can differ in number. The additional

state 4 = one seta often sits a little sideways on the inner margin

state 5 = none of the lobe. We have no preconceived notion as to

is those This character constant within taxa wherein , what might be the derived situation. We suspect,

populations were used for species descriptions, thus however, that the most reduced situation (5 setae)

it seems significant at least for differentiation of reflects the derived state here.

species. This feature is often mentionedonly in the

‘remarks’ section of species descriptions as one that 10. Maxilla 1, palp size

differs between species. The condition in which the state 0 = larger than outer lobe

lobe is fringed with numerous hairs seems to point state 1 = subequal to outer lobe

to an original state. Decidedly more generalized state 2 = smaller than outer lobe

such as elaborate forms, Euphausiacea, have many setae on The need for handling of larger food items

the inner lobe of the maxillule & Walo- short ( Maas may require a long palp (0) as opposed to a

szek, 2001) palp (2). Assumption of the interstitial modeof life,

generally supposed to be a secondary habitat choice

7. Maxilla outer lobe, inner in the evolution of have induced 1 seta (Figs. 5a, 6j, m) Amphipoda, may

state 0 = simple the loss of importance for a maxillular palp.

state I = bifid

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 47

fog- 2. SEM of: b - lateral view of frontal Ingolfiella ischilana, paratype. photographs a, habitus; c, a well-developed cephalic margin

lobe, d - lobe e, seen from outside and inside (removal of first antenna), cuticular ’’hinge” visible (e, right arrow).

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 48 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

3. - Fig. Ingolfiella putealis, paratype. SEM photographs of: a c, e,f, views from different angles on a small cephalic lobe. Photo c reflects best the the possible ‘closing off function of the lobe, a similar fit is made by tiny rostrum (d).

4. SEM of: - views from different the fused Fig. Metaingolflella mirabilis, paratype. photographs a e, angles on partially head region

of third = (arrows); f, cephalic lobe; g, ramus pleopod; h, dactylus of first gnathopod(tip form artefactually distorted). Ceph cephalic

segments (cephalon and first thoracic somite with maxilliped attached). Segm 1 = first body segment with first pereiopod attached.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 49

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 50 R. Vonk & F. R. Schram A phylogenetic and biogeographic analysis

the wherein the 11. Maxilla 1, palp setae out-group share primitive state

= ‘numerous’ state 0 four setae or more setation displays the condition (0). The

= all share state 1 three (Fig. 6j, m) Ingolfiellidae the derived condition of a

state 2 = two reduction to 1 or 2 setae per palp segment (1).

state 3 = one

state 4 = absent 16. Maxilliped, lateral propodal setae

0 = The number of palp setae is a constant feature in state present

those cases where larger numbers of individual state 1 = absent

There- The of this of is recorded in specimens of a species where studied by us. presence row setae

fore, it is a useful diagnostic character. A decrease only a few instances.

in the number of these setae is noted for what is

seen in the out-group taxa. 17. Maxilliped, dactyl claw

state 0 = absent

12. Maxilla 2, setae number outer lobe state 1 = single setae

= limb with 2 = robust state 0 large complex many setae state spine

state 1 = five setae, or more state 3 = spine with flanking setae

state 2 = four setae (Fig. 6k) state 4 = falcate (Fig. 5e)

state 3 = three setae There is considerable variety in the form and num-

= ber of setae the ofthe state 4 two setae on apex maxilliped palp. No

2 limb be The maxilla is a greatly reduced with little, apparent difference in function can ascribed to aside from setal number, to distinguish variation. forms with one or more spines at the apex. One

might be tempted to speculate that blunt and strong

13. Maxilliped basis spines may assist in heavy food particle holding.

= state 0 free and separate However, without functional studies of live mate-

1 = state fused base proximally rial such speculations are of only anecdotal inter-

Free and separate bases of the maxilliped suggests est.

a condition close to the plesiomorph situation, e.g.,

(0) where the maxillipeds are a pair of thoracic 18. Second thoracic segment (= first pereionite)

in 0 = walking limbs. The basis is free and separate state free

mictaceans and in Metaingolfiella (Ruffo, 1968, fig state 1 = fused to cephalon

II, 3) The situation in the family Metaingolfiellidae is

different from that seen in all other species in the

14. Maxilliped lobes Ingolfiellidea and in the chosen members of the

state 0 = with basal lobe only out-groups. Although the half-fusion ofthe cephalon

1 = and state with basal and ischial lobes is an autapomorphy thus phylogenetically un-

Ischial lobes informative in this this on the maxilliped are lacking in Ingol- analysis, character cannot

and fiellidea and in Mictacea (0). Other members of be ignored deserves a place in the matrix to

the out-groups we employed, the bogidiellids and emphasise its peculiarity. Another type of partial the of the pseudoingolfiellids, do have ischial lobes (1). fusion cephalon and first pereionite can be

So this character adds observed in no information to in-group Caprogammarus gurjanovae Kudrja-

polarization of character states. It merely shows schov & Vassilenko, 1966 (in Takeuchi & Ishimaru,

the convergent development of pseudoingolfiellids 1991). This particular member of the suborder Ca-

as compared to Ingolfiellidea. prellidea also exhibits a partial fusion ofthe cephalon

and first pereionite but on the ventral side of the

15. Maxilliped, medial palp setation segment, while in Metaingolfiella the fusion starts

0 = state numerous from the dorsal side (Figs. 4a-e). Flowever, the

= state 1 1 or 2 of the suture is similar in the two per segment position genera, Only Metaingolfiella and the other members of the and corresponds with the position of first gnatho-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 51

P'g- 5. male 2,6 left side b, maxilla 2 left mandible d, mandible Trogloleleupia eggerti, mm: a, maxilla 1, (scale B); (B); c, (A); right

(A) e, of apex female 2.3 1 T. male, 8,5 mm: 2 h, male 12 maxilliped (B); f, mm, pleopod (B). g, leleupi gnathopod (B); mm, 2 gnathopod (B); i, I 1 male 12 antenna 1 with pleopod (B); j, female 7 mm, pleopod (B); k, mm, accessory flagellum and aesthetascs (B) I, specimen of 7 aberrant unknown sex, from a broken pleon, urosome with exceptionally long uropod 3; m, female, mm, gnathopod 2 = = = with 2 teeth of 3 = molar i! ol dactylus instead (C); mp peg; inner lobe; outer lobe, pas palmar angle seta.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 52 R. Vonk & F.R. Schram A phylogenetic and biogeographic analysis

which from their = subchelate pods, are shifted anteriorly more state 1

typical location more posteriad. state 2 = carpo-subchelate

Since all ingolfiellideans have the carpo-subchelate

19. Pereional segments state (2) this character gives no information of

0 = within the state deeper than long generic or specific relationships subor-

state 1 = subrectangular der. The feature is not unique either because, as is

2 = & Poore state elongate remarked by Lowry (1989), it is also seen

This character the overall in unrelated the serves to capture body relatively groups such as the Parda-

habitus This be feature and (Fig. 2a). may an important liscidae, Aoridae, Corophiidae, Leucothoidae,

in the habitat Life in finer the have also regard to requirements. Hyperiidea. Ischyroceridae carpo-

sand & interstices calls for a “worm-like” appearance. chelate second gnathopods (Lowry Springthorpe,

2001) Nevertheless, this feature is scored here be-

20. Coxal plates cause the out-group does show states 0 and 1.

0 = state not developed, or weakly so

state 1 = robust 23. Gnathopod 1, dactyl

Part of the but not Mictocaris exhibits state 0 = no out-group, , gnathopod

the robust development of the coxal plates (1) more state I = simple

typical of amphipods as a whole. It is a unique state 2 = serrate

character of the Ingolfiellidea that the coxal plates state 3 = blade-like

are rudimentary (0). For the Ingolfiellidea as such state 4 = spines

it is an uninformative state. Coxal plates may have The simple state is reflected in a smooth posterior arisen separately on different occasions. For instance, margin ofthe dactylus. This is encountered in Meta-

in in they are present both amphipods and isopods ingolfiella and Stygobarnardia but also Ingolfiella but form in this case one of the few specific char- littoralis (Fig. 6a). The serrate state (Fig. 7a) is acters common to both groups (Siewing, 1963: p.96) exemplified by I. abyssi. Blade-like forms are ob-

served in two species of Trogloleleupia and in Ingol-

Lenticular britannica. the the 21. organs fiella Here spines on posterior

0 = absent are broad and flattenend another of state margin or, way

state 1 = incipient interpreting, the serrated margin has become more

state 2 = well developed widely interspaced. An intermediate form of this

is 2 This a feature encountered, within Ingolfiellidea, is shown in Fig. 5h, on the gnathopod of Troglo-

in the African The in only species of trogloleleupians. leleupia leleupi. When the serrations are broad

lenticular these organs are distinctive, semi-transparent, contrast to the space between, would be in-

The fourth “windows” of cuticle on the side of the segments terpreted as blades. state, spines, is scored of the pereion and pleon (Griffiths, 1989). Their for species with clearly distinguishable rounded

resemble function is unknown. They in some re- spines protruding from the margin.

spects the foramen ovale, a similar transparent area

of cuticle the 24. 2 on heads of the syncarid genus Alla- Gnathopod carpus, palmar angle spines

naspides another meiofaunal crustacean. In Pseu- state 0 = unspecialized

doingolfiella there is mention of “hyaline spots” state 1 = with elongate spine, not angulate but in a different location. Noodt 2 = (1959, p. 203) state pedicillate

remarks on chilensis : “todos los 3 = with Pseudingolfiella state protruding angulate process, spine

toracicos segmentos fibres poseen entre la insercion The palmar angle spines are often used in amphi- de las extremidades un punto circular hialino bien pod taxonomy as an important character in helping

delimitado de funcion desconocida”. determining differences between species within one

These often robust and genus. spines are quite placed 22. Gnathopods at the end of where the tip of the dactylus reaches

state 0 = simpe limb the inner margin of the propodus/carpus. In some

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 53

f'ig- 6. Ingotfiella littoralis, holotype: a, gnathopod 1 ( scale D); b, gnathopod 2 (D); c, pereiopod 3 (D); d, pereiopod 4 (D); e,

5 female pereiopod (D);, f, uropod 2 (D); I. tabularis., female 2.1 mm, mandible(A); h, male 1.9 mm, mandible (A); i, 2.1 mm, ,, , g. , OJ mandible (A); maxilla female, mandible maxilla 1 j, 1 (A); k, male 1.9 mm, maxilla 2 (A); 1, I. quadridentata. paratype, (A); m, (A); n. I. = = margaritae. male = molar il = inner ol = outer lobe, seta; da holotype, mandible(A); mp peg; lobe; pas palmar angle dactylus; cs = cup shaped spine.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 54 R. Vonk & F.R. Schram - A phylogenetic and hiogeographic analysis

used the animal cases the palm makes an angle here as in Troglole- grip when the claspers are to pull

these teeth and leupia leleupi (state 3, angulate, Fig. 5h). When forward. In some species, are broad

into In this seta is placed on a small distinct pedestal we formed blades. other species, these struc-

call this pedicillate (state 2, not illustrated). In the tures are more rounded and form tooth-like out-

the case of a straight palmar margin line and a rela- growths of cuticle. The number of the struc-

tively long seta this is scored elongate (state 1, Fig. tures varies and is also incorporated as a character

6b), and in the unspecialized state (0) the spine is state. relatively short (Fig. 7b) or not present.

28. Gnathopod 2, dactyl tip

25. Gnathopod 2, carpal saw 0 = simple (thin moderate)

state 0 = absent 1 = simple and thick

= state 1 = serrate (Fig. 6b) 2 long and thin

state 2 = dulled (Fig. 5h) 3 = long and thick

= the state 3 setose brush The tip of dactyl could be perceived as a com-

= and the state 4 finely serrate bination of a proximal “dactylar” part dis-

found this feature to be consistent in tal a is We very a “ungular” part, although suture not easily

sizeable population example in at least one popu- observed. Some species, like Ingolfiella canariensis

lation of Ingolfiella canariensis. We examined more and I. similis, have a visible division (Fig. 7b); other

than 50 individuals collected from beach intersti- species do not. The length and thickness of this tip

tia on the Canary Islands, and all specimens exhib- is variable and can be scored.

ited the serrate state (I) (Vonk & Sanchez, 1991).

29. Gnathopod 2, size

26. Gnathopod 2, distal propodus form 0 = subequal to P3

state 0 = unmodified 1 = larger

= state I = tooth-like 2 smaller

= tool state 2 finger-like The second gnathopod is a powerful for an

= robust than the state 3 blade-like ingolfiellid. It is more first gnatho-

that is often The propodus in Ingolfiella is typically smaller than pod, which has a carpus less broad

the and is where and more 7a, b; 8a, The size of carpus positioned at a place nor- pointed (Figs. b).

mally the dactylus is attached, i.e., just distal to the second gnathopod relative to the third pereio-

of the the sub-flexure. This article displays a spur-like pod is measured by summing the lengths

in form from five of both process at its distal end that varies podomeres appendages.

pointed, to blunt, to blade-like, to unmodified. The

propodus in Ingolfiella littoralis shows the unmodi- 1 30. Gnathopod 2, palm

= Tied state (Fig. 6b, arrow), Trogloleleupia leleupi 0 transverse

has the finger-like condition (Fig. 5h, arrow) and 1 = oblique

Ingolfiella abyssi the tooth-like (Fig. 7b, arrow). The form of the carpus is either robustly triangular

with a short palmar margin, “transversely” cut when

27. Gnathopod 2, dactyl teeth seen from the side, as in I. littoralis (Fig. 6b), or

0 = absent elongate with a faint sloping, oblique palmar side

1 = 3 teeth (Fig. 6b) as in I. abyssi (Fig. 7b).

2 = 3 blades (Fig. 5h)

3 = 4 teeth 31. Pereiopods 3 and 4, claw

4 = 4 blades (Fig. 7b) 0 = absent

We observed differences in the form of the teeth 1 = simple

inner = lining the margin of the dactyl of the gnatho- 2 dentate or bifid

pods. These thorn-like structures probably help in Spiny structures on the termini of the third and fourth

and/or or the be used for better Most securing prey mates, to strengthen pereiopod may grip. spe-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 55 Contributions to Zoology, 72 (I) 2003

F ‘g. 7. Ingolfiella abyssi. holotype: a, gnathopod I ( scale D); b, gnathopod 2 (D); antenna 1 (D); d, pereiopod 3 (D); e, pereiopod _ . c, (D), f, pereiopod 6 7 1 2 3 da = (D); g, pereiopod (D); h, plcopod (D); i, pleopod (D); j, pleopod (D); k, uropod 1, (D); dactylus; pas ~ P‘dinar angle seta.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 56 R. Vonk & F.R. Schram A phylogenetic and biogeographic analysis

cies in Ingolfiella have the dentate or bifid state 36. Pleopod form

and few 0 = biramous (Fig. 7d), only a possess a simple spine.

All species in Trogloleleupia, the large African cave 1 = uniramous, narrow (Fig. 5i)

species, have the simple state. 2 = uniramous, short fin

3 = uniramous, long fin (Fig. 5f)

32. Pereiopods 3 to 7, dactyl form Reductions of various sorts can be observed through-

0 = similar out the family Ingolfiellidae. Truly interstitial life

1 = dissimilar of a certain mode might induce short appendages

A feature in in- the end of common gammaridean amphipods on rear, or pleonal, a wriggling body

in worm-like volves the different positioning of the third and that moves a fashion. In this vein, the

fourth pcreiopod in relation to other pereiopods, oceanic mud-dwelling species and the terrestrial

the and seventh. Pereio- c.g., fifth, the sixth, the cave-pool species would tend to have longer pleo-

unit’ beach pods 1 to 4 form an ‘embryological (Dahl, pods than the and riverine taxa that occupy

would functional/mor- interstitial between sand 1977), or, as we interpret, a open spaces grains.

phological unit. However, the form of the dactylus also can differ between those of P3 and P4 on the 37. Uropod 1, rami

one hand, and those of P5-P7 on the other. For 0 = free

in = instance I. littoralis the forms ofthe dactylus on 1 fused at peduncle

P3 P4 In and (Fig. 6c, d) are dissimilar (state 1) from Metaingolfiella, the aberrant species from a deep

P5 (Fig. 6e). well in Italy, the rami are fused at the peduncle (1)

In all other species in this analysis, the rami are

33. Pereiopods 5-7, dactyli free (0). This character provides no information to

0 = with claw the but is maintained to stress the phylogeny pecu-

1 = without claw liar situation in Metaingolfiella.

The dactylus has a separate claw in for instance

Ingolfiella abyssi Fig 7. d-g). Other species, like 38. Uropod 1 and 2, size comparison

lack such = 1 I. littoralis (Fig. 6e) a claw. 0 uropods and 2 not present

1 = subequal

34. Pereiopods 3-7, dactyl ends or termini 2 = u2 larger than u 1

0 = not produced 3 = u 1 larger than u2

1 = with This character measures the size of the first and spur (Fig. 6c)

2 = with seta (Fig. 7d) second uropods relative to each other. No clear

and This character involves an added bit of decoration pattern emerges although we consider these

to the distal aspect of the dactyli. The spur is a types of characters susceptible to allometric change,

robust is alternative in spike-like process and opposed to a more there is no dealing with these append-

flexible extension. and molt influence setose ages. Age stage might perhaps

the relative size of the rami, but it is impossible to

35. Pleopods 1 - 3, female effectively take this into account. Our reasoning to

= 0 present include this feature here is that if there is a clear

1 = it due absent pattern, then we take the risk that is to coin-

There is a trend in ingolfiellids toward reducing cidence; and if there is no pattern, then nothing is

the pleopods to a minimum number. The absence lost. of all the pleopods would be the ultimate condition

in this regard. The problem with assessing this 39. Uropod 1 and 2, total length

feature = adequately is that we do not always have 0 no such uropods present

sufficient of both = samples sexes for some species 1 long of ingolfiellids. In males, the first pleopods always 2 = short

seem to be present. See comments vis-a-vis character 38

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 57

40. inner in with better it Uropod 1, versus outer ramus length point the future, collections, may

0 = in equal length become feasible to assess these features a phy-

1 = than halfofinner outer ramus more ramus length logenetic context across all species.

2 = outerramus less than halfof innerramus length

The variation of both rami of is in length uropod 1 2 element a. Gnathopod male, reverse on carpus not No obvious function be readily explained. can 0 = absent

ascribed It to this difference. is possible that the 1 = present outer which is in line with the underside of ramus, At the end of the palmar margin, in the region of the body, should be longer to protect the inner ra- the large palmar setae delimiting the palm, an out-

mus and the urosome from or to interact damage of cuticular tissue often growth irregular appears actively with other appendages (as was observed in older males (Fig. 6h) by Spooner, 1961, in grooming).

b. Uropod 2 male, baso-facial spine 41. Uropod 3 0 = absent 0 = biramous 1 = present 1 = uniramous In some species, a conspicuous, often somewhat This character in mem- appears Pseudoingolfiella, a hooked, is on the basal part of the ber spine present of the out-group but with a uniramous third peduncle in the male. In other species, this is ab- uropod (1), and emphasises the most interesting role sent, no intermediate situation has been reported. of such intermediate taxa.

c. Oostegites 42. Uropod 3 fusion

= 0 0 present = right and left unfused

1 = absent 1 = fused peduncle All species in Ingolfiellidea would, by definition, Only in Metaingolfiella are the peduncles fused to

each be expected to have broodplates. But only few could other. The tendency towards fusing of ele-

in this is three-fold also be scored. ments species now, given

the fusions already noted in the head and the rami

of uropod 1. Tendency in this species seems to be Cladistic analysis towards fusion of adjacent elements into large units.

43. Using PAUP 4.0 blO (Altivec) we analysed the Tel son form (Fig. 51) character matrix that entered into 0 = (Table we well developed 1)

1 = MacClade 4. An heuristic search with unordered medium, fleshy

2 = and characters resulted in 4 trees with short, flat unweighted

3 = a of 261 When the multistate short, fleshy tree length steps.

4 = treated short, bifurcate characters are as polymorphism rather than

A the 295 but the sturdy, well-developed telson is consideredplesio- uncertainty tree lengthens to steps,

the same. We chose morphic with reference to the Amphipoda (Barnard topology stays one tree (Fig.

& the Karaman, 1983). 8) upon which to optimize obtained character

transformations. The other three trees differed only

The in following characters were also examined. slightly out-group topology or character optimi-

They were not used in zation in that the clade had different our final cladistic analysis, Ingolfiella a

however, because within the brackish water there were too many question grouping species arrange- marks concerning them in the matrix. They all in- ment (Fig. 8, insert). However, these trees could

volve sexual have well differentiation, and because of the rarity equally served the purpose. of most The and selected ingolfiellideans it is not always possible to analysis used Mictacea gam- have adequate that contain both sample sizes sexes. maridean amphipods as out-groups. The suitabil- We list them here, nevertheless, because at some ity of Mictocaris halope as an out-group taxon arises

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 58 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

its - from comparable habitat requirements marine placed capital letters to focus attention on the most

caves on Bermuda - and also a similarity in form interesting branches of the cladogram. These

and position of oostegites that are of “gammaridean branches display good apomorphies, i.e., with high

For lack of amphipod type” (Just & Poore, 1988). In addition, consistency indexes. instance, the an

ischial lobe the in the lobe on the frontal margin of the cephalon is (character 14) on maxilliped

comparable to the equally enigmatic lobes in Ingol- Mictocaris halope results in a strong emphasis by

ftellidea. also the on the of this lobe in Mictocaris possesses reduced pleo- program presence Bogidiella and This is also illus- pods. The plesiomorphic state of cephalic and tho- Pseudingolfiella. example

trative of how such must be in racal appendages serve as a basis for polarizing changes interpreted relative The mictaceans supposedly homologous features in the other taxa a way. are not necessarily

a close sister to the and included in the analysis. group bogidiellids pseud- ingolfiellids (node A). We lack in fact countless The gammaridean amphipods Bogidiella Hertzog,

numbers of relevant to 1933, and Pseudingolfiella Noodt, 1965, have closer species amphipod history it that never made through to the and we affinities to ingolfiellideans than Mictocaris. Bogi- present,

undoubtedly lack many deep groundwater forms diella has reduced pleopods and a strictly stygobiont that could enlighten us on the origins and early life cycle, with most representatives living in fresh evolution of the We wish group. merely to say, water but also a few in the brackish and marine

before we go on to interpret character state changes, environment. Their body plan is certainly less re- that these taxa and these charac- thus holds given particular duced, and potentially more plesiomor- ters, and provided one watches out for specific phies, than that of the ingolfiellideans. The overall then the habitus assumptions, particular phylogenetic pat- of Pseudingolfiella earned this genus its terns we obtained seem to prevail. name. Pseudingolfiella has much reduced pleopods, At the base of the clade, we note in ingolfiellidean as Ingolfiellidae. Two species are known: Pseud- characterized the Metaingolfiella, by uniquely par- ingolfiella chilensis (Noodt, 1959) and Pseudin- tially fused second thoracic segment into the head, golfiella soyeri Coineau, 1977. We used both of and the fusion of elements in uropod 1 rami and in them the out-group as bridging forms between uropod 3 peduncles. (This last feature is interest- and Ingolficllidea Mictocaris and as stand-ins for , ing in regard to the caudal furca/rami situation in the gammaridean amphipods. other crustaceans, Schram & Koenemann, in lit.). We have considered using as out-groups the other Nevertheless, these autapomorphies are associated two remaining, exclusively marine, suborders in with features that insures this many plesiomorphic the Amphipoda: Caprellidca and Hyperiidea. How- species remains at the base of the ingolfiellidean in ever, many characters our matrix would have clade. been left unscored. The adaptations found in caprel- Stygobarnardia (node B) regains a vestigial man- lids and hyperiids, which relate to their very dif- dibular palp in this scenario. It was lost in the tran- ferent of made them way living, an unlikely choice sition from the out-group to the in-group (node A). for comparison. Caprellids cling to sea weeds and such character While reversals appearto pose prob- lower metazoans in the sublittoral and they zone, lems for diagnosing taxa, they can perhaps best be have a modified are strongly pleon. Hyperiids pe- understood in terms of paedomorphic shifts of de- lagic with a highly modified head region. velopmental timing of events.

The “trogloleleupians” emerge as paraphyletic.

of The members a small clade (node C), made up Results of T. eggerti, T. dracospiritus and T. leleupi, share

weakly developed lenticular organs, while the other The cladogram (Fig. 8) reveals a far from random two “trogloleleupian” species are set off by having array of The species. out-group, Mictocaris, Bogi- either lenticular strongly developed organs, or a diella and the two of species Pseudingolfiella re- , differently implanted palmar angle seta. Two new mained from the apart in-group whether or not we genera are necessary to recognize their isolated enforced an out-group constraint in PAUP. We have locations on the tree.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 59

Fig- 8. Tree number 1 (rooted by using user-specified outgroup): Tree length = 261, Consistency index, (Cl) = 0.3511, Homoplasy

index = (HI) 0.6489, Cl excluding uninformative characters = 0.3436, HI excluding uninformative characters = 0.6564, Retention

nidex (Rl) = 0.5550, Rescaled index = 0.1949. on several nodes consistency (RC) Capitals (A-G) represent groups ( see text). Species

Ingolfiella without a symbol for subgeneric classification entered the literature after 1989. Insert: alternative topology within the brackish water clade.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access - 60 R. Vonk & F.R. Schram A phylogenetic and biogeographic analysis

Ingolfiella is retained as a distinct genus despite Systematics its wide habitatand biogeographic spread. The short-

of the stands out without ening uropod 1 as an apomorphy. Of course, one can diagnose any taxon

the This character state is stable throughout genus, benefit of a rigorous cladistic phylogenetic analy-

Other the of no alternating of states is encountered. changes sis. However, ideally, taxonomy a group

node the transitions from defining the Ingolfiella are should reflect the phylogeny. In consequence, now blades the of the in for the to spines on margin dactylus having performed first time a cladistic analy-

number of gnathopod 2, and the diminishing crene- sis of the entire suborder Ingolfiellidea, we can go lations the reversals on palmar margin (although back and present taxonomic diagnoses that emerge

scattered to the absent state are seen, throughout naturally from the phylogeny. We provide the su- the Ingolfiella clade). pra-specific taxa of the suborder below with such

A coastal Mediterranean clade (Fig. 8, node D) diagnoses (diagnostic apomorphies). For a world is separated on the basis of losing one seta on the catalogue and bibliography of the Ingolfiellidea we

rela- www.uba.uva.nl/ctz inner lobe of maxilla 1. These setae are sparse refer the reader to tive to what is common in gammaridean amphi-

three pods. Reverting from four to setae would seem

reduce the of that limb for food to power handling Taxonomic diagnoses derivedfrom cladistic particles. analysis

The transition to the brackish water grade (Fig.

is 8, between nodes E and F) defined by a single Suborder Ingolfiellidea Hansen, 1903 character state shift: the size of gnathopod 2 going from smaller than the of length pereiopod 3, to Diagnosis. - Mandible palp rudimentary or absent,

this molar maxilla 1 subequal to length. process vestigial peg or spine; outer

marine clade be Finally, a can recognized (node lobe outer setae generally dentate; maxilliped with

the basis of the F) on gaining one tooth/spine on only proximal part of basis fused; pereional seg- dactylar margin of gnathopod 2, but reversals are ments subrectangular to elongate (not deeper than

further in the clade. a possible up Furthermore, long); gnathopods carpo-subchelate; gnathopod 2 process on the tip of the propod reverts to an un- palmar angle setae variable (seldom simple), dac- modified this node but is reversed in thinner than state on again tyl tip elongate (usually thick); uro-

the At- the Australian species, South-African, the pod 2 generally larger than uropod I or subequal. lantic and Caribbean Ocean, a species. Remarks. - Of these characters, the features con-

four We have omitted from our analysis species; cerning the mandibleand the carpo-subchelate nature

littoralis and Ingolfiella abyssi, I. atlantisi, I. I. of the gnathopods are robust, with consistency in-

character for kapuri. The many unscoreable states dexes of 1. Hansen (1903) utilized two of these these species clouded the initial analyses that we features to erect his superorder, and we can now

When ran. we removed one species after the other see from our analysis that in fact these are robust from the heuristic searches, the number of trees features. However, these exclude the plesiomorphic became fewer and the of the branching pattern re- features one could use. Thus, we have some addi- maining taxa began to stabilize. We would hope tional arguments to supply towards justifying Ingol- that some additional material of these will fiellidea natural day taxa as a taxon, e.g., vis-a-vis the argu- become available that will allow more characters ments of Lowry & Poore (1989) and Dahl (1977). be to scored for these species. Furthermore, it is The state of the maxilla I outer lobe outer setae our desire their that by laying out the characters and and the dactyl tip of the second gnathopod are also

alternative in the detail that = possible states we have, useful in diagnosing this suborder (ci’s 0.4 or that future describers of ingolfiellidean species will higher). However, the other features, while help- become more focussed what in on descriptors they ing to define the clade on our analysis, never- should seek to elicit from the of their For study speci- theless, are rather homoplastic characters. in- mens than has been the case to this up point. stance, the palmar angle setae of the second gnatho-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 61

maxilla pod is quite variable displaying elongate, pedicillate, Diagnosis. -Mandible palp vestigial; 1 outer

forms 2 car- angulate and these are diagnostic only in so lobe outer setae typically bifid; gnathopod

ends far as they are seldom simple (see character dis- pal saw typically serrate; dactyl on pereio-

3 1 and cussion). The relative sizes of the uropod 1 to uro- pods to 7 produced as a seta; uropod uro-

2 pod 2 are even more homoplastic. pod subequal.

Remarks. - Of these features, only the state of

feature this Family Metaingolfiellidae Ruffo, 1969 the mandibularpalp is a unique for genus.

other features above Metaingolfiella Ruffo, 1969 The noted also appear amongst

various species of the genus Ingolfiella.

Type species. - Metaingolfiella mirabilis (by mono- typy) Genus Trogloleleupia Ruffo, 1974

Diagnosis. -Ocular lobes developed; maxilla I with 1951. single inner lobe seta, 4 or more palp setae; max- Type species. Ingolfiella leleupi Ruffo,

- 1 with 4 illiped lateral propodal setae present; second tho- Diagnosis. Maxilla palp or more setae;

fused lenticular at least racic segment partially to cephalon; gnathopod organs incipiently developed; in 2 carpal saw as setose brush; uropod 1 rami fused gnathopod 1 dactyl serrate; telson medium size

to peduncle; uropod 3 fused peduncles. and fleshy.

= - the state of the Remarks. - The most effective characters (ci Remarks. Of these features,

lenticular is critical. 1) towards defining this monotypic family are the organs In Trogloleleupia eggerti

partial fusion of the second thoracic segment to the and T. leleupi they are only incipiently developed, well This head and the two features involving the uropods. but in T. dracospiritus they are developed.

In here featureof the char- contrast, the other characters mentioned is the most important diagnostic

acters also display some degree of homoplasy. above, although well-developed organs

It occur in Paraleleupia gobabis. is noteworthy to

Family Ingolfiellidae Hansen, 1903 mention that in T. dracospiritus we also see the and bifurcate unique appearance of a short telson,

T. Diagnosis. - Antennuleflagellum moderate in size, although in our analysis (Fig. 8) dracospiritus

is related T. than T. i-e., more thanhalf the length of the basal peducular more closely to leleupi to eggerti.

of 3 A former definition of the segment, accessory flagellum typically seg- genus Trogloleleupia

ments; maxilla 1 outer lobe spines generally 6 in by Ruffo (1964) was based on T. leleupi and T. their number, palp most often subequal to outer lobe; eggerti and mentions large body length, pres-

with 1 2 thoracic 3- maxilliped medial palp setation only or ence of lenticular organs on segments

setae than and of the per segment; pleopods more often not 10, long pereiopods uropods, peducle

as short fins; telson short and fleshy; uropod 1 typi- second uropod with numerous rows of setae, and

with first cally outer ramus less than half the inner ra- dimorphic pleopods.

mus.

Remarks. - that of new Of these features, only one, Proleleupia genus the medial palp setation, is robust (ci = 1). In ad-

dition, the size of the antennular flagellum and the Type species. - Trogloleleupia midicarpus Griffiths, number of outer lobe spines on the maxilla 1 have 1991.

an - acceptable consistency index (ci > 0.5). All the Diagnosis. Maxilla 1 inner lobe setae 3 in num-

other the 1 features mentioned above are rather homoplas- ber, palp larger than outer lobe; gnathopod tic (ci < 0.5). dactyl with innermargin spines blade-like; gnatho-

pod 2 carpal saw absent.

Genus - A for Stygobarnardia Ruffo, 1985 Remarks. separate genus, Proleleupia,

■what was known as Trogloleleupia nudicarpus Grif- Type species. Stygobarnardia caprellinoides fiths, 1991 is required by our cladistic analysis,

Ruffo, 1985 (by monotypy) wherein the formerly large-bodied genus Troglole-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 62 R. Vonk & F.R. Schram - A phylogenetic and hiogeographic analysis

leupia Ruffo, 1974 emerged as a paraphyletic taxon. pod 2 stand out as very diagnostic for the species

of the within the These characters The core species trogloleleupian group, con- ingolfiellideans. are quite

the for the alternate taining type species Trogloleleupia, T. leleupi stable throughout genus: no states

(Ruffo, 1951), is part of a fairly distinct clade, but concerning the uropods are encountered, and di-

of the five in the variants of the two species formerly placed ge- verse palmar angle seta can be seen

nus Trogloleleupia in fact sort separately from this except the pedicillate form. In addition, species of

nudi- have serrate core species group. In addition, Proleleupia Ingolfiella generally a carpal saw on carpus is the only trogloleleupian that lacks len- gnathopod 2, exceptions being I. cottarellii, in which ticular organs, a negative feature to be sure, but it is uniquely finely serrate, and I. beatricis, I.

nonetheless one that separates it from all the other macedonica, I. xarifae, I.fuscina, and I. grandispina,

large-bodied trogloleleupians. in which it is absent.

The genus Ingolfiella is retained here as a single,

Paraleleupia new genus undivided taxon, even though it is world wide in

distribution and can be found in localities ranging

Type species. - Trogloleleupia gobabis Griffiths, from fresh groundwater, through diverse brackish,

1989 to deep-water ocean habitats. Given this wide dis-

- Diagnosis. Pereional segments sub-rectangular; tribution and variety of habitats it is instructive to

lenticular well 2 examine several the for organs developed; gnathopod dactyl points on cladogram Ingol-

4 with teeth developed as blades, dactyl tip simple; pleo- fiella further comment as to character state

pods as long fins; uropod 1 and uropod 2 subequal. changes.

- A tor Remarks. separate genus, Paraleleupia, A coastal Mediterranean clade (Fig. 8, Node D)

what was known as Trogloleleupia gobabis Griffiths, is separated on the basis of loss of one seta on the

1989 is required by ourcladistic analysis, wherein, outer lobe of the maxilla 2. These setae are reduced as mentioned earlier the large-bodied genus Troglo- relative to what is common in gammaridean am- leleupia Ruffo, 1963 emerged as a paraphyletic phipods. Reverting from 4 setae to 3 setae would

of this limb taxon. Its well-developed lenticular organs clearly seem to reduce the ability to handle characterize Paraleleupia gobabis, a feature shared food particles. This may appear as a “subtle” fea-

T. In only by dracospritus. addition, the simple dactyl ture, nevertheless the members of this clade ex-

of tip on the second gnathopod is a feature seen also hibit a relatively high degree geographic prox- in T. eggerti but no place else amongst all the ingol- imity. fiellideans. for Prole- from freshwater By erecting separate genera The change over point to brackish leupia nudicarpus and Paraleleupia gobabis we habitats (Fig. 8, Node E) is also defined by a single

of taxonomic rela- character state transition: the size of 2 effectively set up an hypothesis gnathopod

in of the from smaller than of tionships this part tree. We can test this shifts being the length pereio- hypothesis either with discovery offuture ‘troglole- pod 3 to being subequal to the latter. It again is

accumulation of difficult conceive of leupians’, or sequence data, or to any purported selective further into of of such a feature. insight gross features morphology. advantage Nevertheless, this

node marks a distinct habitat shift within the ge-

1903. nus. Another critical habitat shift into Ingolfiella Hansen, occurs pure

marine waters (Fig. 8, Node F) and corresponds

Type species. - Ingolfiella abyssi Hansen, 1903. with changes in aspects of the second gnathopod.

- At Diagnosis. Gnathopod I dactyl either as a simple this point we note an alteration of ornament on spine or serrate, gnathopod 2 palmar angle seta not the dactylar margin of gnathopod 2, going from 3 pedicillate, carpal saw typically serrate; dactyl ends blade-like processes to 3-4 teeth. Furthermore, we

3 on pereiopods through 7 typically produced as a also note increasing variability on the distal propodus

spur; uropods 1 and 2 short. of gnathopod 2. Most freshwater Ingolfiella dis-

Remarks. - The nature of the short uropods and play blade-like distal propodi. Within brackish ha-

the lack of a pedicillate palmar angle seta on gnatho- bitats we also see a finger-like form appearing. In

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 63

the marine an unmodified and realm, propodus reap- geographically ecologically, genus as Ingolfiella

pears, a structural form characteristic of taxa typi- currently represents can be revisited.

outside the cally genus.

What surprises us about the above points is that Additional descriptions

while the habitat or geographic shifts are rather

striking, the anatomical changes at these points are Re-examination of type and other material in light

subtle in the extreme. of the cladistic analysis allows additions and revi-

Past authors (S. Karaman 1959, Stock 1976b, sions to be made to existing species descriptions.

Ruffo 1970, and Ruffo & Vigna Taglianti 1989),

have made arguments to subdivide the genus Ingol- Metaingolfiella mirabilis Ruffo, 1969 fiella into either least 4a-h more genera, or at subgenera Fig.

- the (see bibliography). As can be seen in Fig. 8, these Material examined. collection of Natural His-

subgenera have little value in light of the cladistic tory Museum Verona.

here. analysis Nevertheless, we do see clades ap- Additional redescription: cephalon fused over about

within the have noted half the lateral side with first pearing genus. We already lower the pereionite.

the Mediterranean clade of the I. beatricis species

group, characterized by a single feature. Another

such clade is the I. macedonica species group (Fig. Trogloleleupia eggerti (Ruffo, 1964)

8, Node G). This clade is characterized by two good Fig. 5a-d

4 features: 1) the claws of pereiopods 3 to are simple in and 3 to 7 lack Material examined. - collection of the Natural His- form, 2) pereiopods dactyls any decoration such as spurs or setae. These are good tory Museum Verona. features at that point in the tree. If characters nicely The allotype (female) designated by Ruffo was

delineating discrete clades would justify the erec- restudied and some more sexual dimorphism is

tion of subgenera in a cladistic context, one might reported. Not only is the form of the first pleopod be tempted to look upon the so-called I. macedonica different - elongate in males and triangular in fe-

We be - species group as a good one. might even males as mentioned by Ruffo, but also the fe-

tempted to this is even more robust male lacks a hooked on the base of the say species group spine pe-

than that of I. beatricis discussed duncle and the species group of the first uropod claviform pro-

above, which only has a single, rather subtle diag- cess on the peduncle of the second uropod. An-

nostic feature. the I. beatricis is other difference is the lack of However, group noteworthy the pal-

while the members of in the second geographically continuous, mar corner process gnathopod, but

the so-called macedonica I. species groupare hardly this occurs in other ingolfiellids too.

so: I. macedonica is found in northwestern Greece, Additional description: I. manni was collected from northern Chile and I. Maxilla 1, (Fig 6 a) palp two-segmented with 4

uspallatae comes from the Andean of distal Outer lobe with high passes setae on segment. 6 serrated

western Argentina. How do we bridge the gap from and combed robust setae. Inner lobe with 5 setae the Balkans to South America? almost evenly distributed along its margin. fhus, at this time we see no justification for Maxilla 2 (Fig 6b), with both lobes having 5 setae,

establishing subgenera, or breaking up Ingolfiella some plumed.

lnt0 separate The Mandible Left side genera. shifting morphological (Fig 6c, d). with pointed pars changes noted amongst the optimized characters molaris and 3 hooked and serrate spines at the base °n the cladogram are too subtle and homoplastic, of the masticatory teeth. Right side with two such ar| d discrete clades are too few and far between. curved hooks.

Wc remain confident, however, that as more taxa Maxilliped (Fig. 5e). Palp with 3 spines apically are a(ldcd to the genus the cladogram will acquire ■and 1 spinule on a small pedestal submarginally

some additional structure. Then at that time, the on the propodal segment. 'ssue ot what to do with 1 such a wide ranging, both Pleopod (Fig. 5f). Triangular fin with a slightly

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 64 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

serrated hind corner margin and 3 tiny spinules on Gnathopod 2 (Fig. 6b), margin of palm serrated.

the lower margin Pereiopod 3,4 (Fig. 6c, d), claws tapering to a forked

end

Trogloleleupia leleupi Ruffo, 1951 Pereiopod 5 (Fig. 6e), claw thick, undifferentiated.

Uropod 2 (Fig. 3g), with three oblique rows ofsetules

Material examined. - British Lusaka Bore- the inner side of and small Museum, on the peduncle a group

9 Institute holes, Zimbabwe, specimen; Zoological of setules on the inner side of the inner ramus. and Museum, University of Hamburg, 15 specimens,

Kivu, Distric de Kindu, Territoire de Kasongo, Ingolfiella tabularis Stock 1977

Grotte de Mwana-Kussu, Congo, 24-X-1954

Material examined. - Zoological Museum of Am-

In the descriptions of Ruffo (1951) and Ingle (1961) sterdam, paratypes, cat. nr. 106.107 the no special emphasis was placed on form of the Additional description; first These have in pleopod. pleopods most ingol- Mandible with (Fig. 6g, h, i), cup shaped spine on fiellids different form in males and females. How- a left mandible. ever, in the many specimens of Trogloleleupia (44 Maxilla 1 (Fig. 6j), with 6 combed and serrate se- + from Belgian Congo and 9 from Lusaka, Zimba- tae of irregular length on outer lobe. Inner lobewith bwe), no distinction could be made between the 1 seta, palp with 3 setae on apex.

sexes. We checked the for such differ- pleopods Maxilla 2 fitted with both (Fig. 6k), sparse setae on ences. Indeed, some of the smaller carry specimens lobes.

the digitiform first pleopods with two setules on

the of while the apex typical males, larger speci- Ingolfiella margaritae Stock 1979 mens have the more spatulate crenelated type as in

pleopod 2 and 3. Material examined. - Zoological Museum of Am- Additional description: sterdam, holotype cat. nr. 106.443 - 444 Antenna I (Fig. 5k) with a 4-segmented flagellum, Additional description: bearing an aesthetasc on each segment. Accessory Mandible(Fig. 6n), with apparently broken offmolar flagellum 3-segmented. Left side could not be discerned. process. or right Gnathopod 2 (Fig. 5g, h) differs in aspect with Three curved spines just below pars incisiva. increasing size, the 8 mm male (6e) has a hyaline

lobe aligning the margin right under the palmar Ingolfiella quadridentata Stock 1979 process but this is not seen in a 12 mm specimen

with The (Fig. 5h), a sculptured palmar process. Material examined. - Zoological Museum of Am- crenelations are more numerous on the palmar mar-

sterdam, paratypes cat. nr. 106.445 - 446 gin. Additional description: Pleopod I (Fig. 5i, j) is long and slender in the with Mandible (Fig. 61) pointed molar process. Three male, having two setae on the apex, the female form small below the incisiva very spinules pars is somewhat triangular and lightly serrate on the 1 with 6 Maxilla (Fig. 6m) has an outer lobe robust hind margin. spines, bifid or serrate. Innerlobe with 1 seta, palp Uropod 3 (Fig. 51) in one instance quite long, with with 3 on apex. five slender its setae on apex.

Ingolfiella littoralis Hansen 1903 Ingolfiella abyssi Hansen 1903

Material examined. - Holotype specimen, Zoologi-

Material examined. - Museum of Copenhagen cal Museum of Copenhagen (ZMUC-CRU-7088) ZMUC-CRU-5030 Additional description: Additional Gnathopod 1 (Fig. 6a), margin of palm smooth, description: Antenna 1 with without structures or serrations (Fig. 7c), 4-segmented flagellum,

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (1) - 2003 65

aesthetascs on the second and fourth the data bearing seg- suggested by at hand. The biogeographic ment. Accessory flagellum 3-segmented. history of the separate clades (Fig. 8, nodes A-G)

Gnathopod 1 and 2 (Fig. 7a, b), with 4 incisions on can be projected onto paleomaps (see Ebach & the inner of the 4 blades. This is margin dactylus, forming Humphries, 2002). a process of straightfor-

2 claw with clear articulation But where Since Gnathopod a between ward inspection. to begin? our dactylus and unguis. cladogram clearly points to an early development

Pereiopods 3-7 (Fig. 7d-g), with bifid claws in p3 of the large fresh water African and Mediterranean and and in it p4 (8d) straight p5-7 (8e, f, g) cave species leads us to look for a paleomap, in

1-3 trian- that would Pleopods (Fig. 7h-j), with more or less this case Triassic, combine both areas gular form. and show a continuous land mass with those areas

1 with Uropod (Fig. 7k), 3 spiniform processes present at some particular point in earth history.

inner Actual fossil (instead of 4 in the original description) on evidence for the existence of the Am- ramus. Outer ramus with a breach on two-thirds of does phipoda not go beyond approximately 40 mil- its lion and length, dividing the ramus in two articles. years ago (Coleman Myers, 2000), when

epigean as well as hypogean forms where trapped

in amber resin. These Cenozoic fossils closely re-

Biogeographic analysis semble the living forms of today, and therefore

would seem to point to a much older amphipod origin

An within accepted approach to elucidating patterns of perhaps one more in line with other groups

historic biogeography would typically begin with the such as the isopods, tanaids, cuma-

either COMPONENT and of a Brooks Parsimony, or Anal- ceans spelaeogriphaceans. Fossils these lat-

ysis. This would require that the distributioninfor- ter groups date back to the Carboniferous (Schram,

mation (Fig. 9) be converted into a supplementary 1981). is matrix, which would then be subjected to further Why it that no Paleozoic or Mesozoic fossils found analysis. Though effective and widely accepted, are among the Amphipoda? The reason is these methods twofold. Either no form exist- pose some conceptual disadvantages. probably recognizable The biogeographic analysis is performed on top of ed, i.e., somehow the amphipods arose from rela- the results of a previous cladistic analysis. This tively recent ancestors, or the amphipods living in

the occurred in imposes a number ofnew equally parsimonious trees Mesozoic (or earlier) habitats that

on top of the supposedly equally parsimonious trees were quite unsuitable for fossilization. These lat-

from the In ter could be base analysis. effect, one accumulates ancestors deep sea, interstitial, and/or

uncertainty on top of uncertainty. cave inhabitants. We assume that the colonization

There of the marine interstitial small benthic are, of course, ways to handle all these by crusta-

alternative but there is ceans was ancient event. A trend be in trees, an additional concep- an can seen tual problem to this approach. This treats the bio- the evolution of worm-like bodies in amphipods,

older in geographic as if it were a which invaded areas ancient times and are history completely sepa-

rate in that and subsequent set of events in the evolution present today areas have been emergent at °f a the group when in fact the spatial and distribu- least since late Cretaceous. This stands in con- b°nal components are an integrated part of a taxon’s trast to forms with rounder bodies, which resemble One their benthic relatives biology. possible solution to this could be to more (Vonk, 1990; Coineau, attach a step matrix to the primary analysis that 2000). would code for all The that possible biogeographic move- present day ingolfiellideans occupy un-

ments within the recorded of the far inland the continents range group in derground waters on are question (Berge, 2000). The biogeography could the large cave lake inhabitants of Africa below then be integrated into the base analysis. This is the equator. Their counterpart, in body length and

* worthy of further micro-environmental exploration within the Amphipoda, requirements, is the one spe- but must be the subject of a separate study. cies of Metaingolfiella, known from a well in Italy. A conceptually simpler approach, however, is Returning to our original question of where to be-

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 66 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

branches.

its

on projected characterstates geographic

Recent

with Cladogram

9.

Fig.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 67

gin and project a common history on a paleomap, mitive forms. Such ideas have been formulated it seems justifiable to search for to a time where earlier(Ruffo, 1951;Leleup, 1955, Siewing, 1963), the African and Mediterranean freshwater forms but Dahl (1977) remarks that a reduction of pleo- could have had The from the is unusual in contact. map Early pods species living in larger bodies

Triassic (Scotese, 1977) reflects such a situation. of subterraneanwaters and thus he maintained that

The Pangaea and Gondwana geography of the Tri- this feature must have been inherited from intersti- assic can be fitted to the branches of those taxa tial ancestors without fully developed pleopods. that group together low in the clade of the ingol- Thereforeit is unlikely, according to Dahl, that these

could fiellidcans (Fig. 10, Fig. 11A). cave forms have been ancestral to the smaller

“Above” these stem taxa on the tree, we encounter interstitial types. However, a detailed description species of Ingolfiella found today in groundwater of Leleup (1955) on the ecology of Trogloleleupia habitats along the North Mediterranean coasts. As leleupi, a large cave-inhabiting ingolfiellidean from

if above, we seek a period after the Triassic, we Congo, reveals that this species does not employ a find in the Jurassic central behavior but rather Early a time when the free swimming moves across Atlantic Ocean and western Mediterranean Sea was the bottom on its side. This suggests the true bot- forming from the older West Tethys Ocean. We tom dwelling nature of ingolfiellideans. Further can the plot taxa that are found nowadays in the speculations on whether small interstitial forms with northern Mediterranean near-coastal areas onto the reduced pleopods transformed back into larger cave northern West Ocean in coast of the Tethys the Early forms with,- secondarily derived, functionally ac-

Jurassic There are tive be further geography (Fig. 11B). excep- pleopods cannot investigated any tions, like Ingolfiella macedonica and especially I. without fossil forms. ischitana in A that occur this same geographic area number of alternative evolutionary scenarios but appear higher in the cladogram (Fig. 10). concerning ingolfiellidean origins now present them- In similar a manner, we find that the grade on selves. First, a marine ancestor could have invaded the cladogram that represents the freshwater and the fresh-water underground environment in the brackish from the Atlantic Is- Late species present day Early Triassic, or even Paleozoic, with con- lands and Caribbean area are easily projected pa- comitant anatomical reductive adaptations taking

onto the northern coast of Late leogeographically place. In some cases, a “rebuilding” of reduced Jurassic Gondwana (Fig. 12A). features (pleopods, third uropods) might have oc- The extremely high sea levels of the late Creta- curred especially when the infestation of the inter-

ceous and further opening of the Atlantic Ocean stitial environment (with its confined spaces) was (Haq et ah, 1987) could have seen an evolution followed by radiation into cave-lake systems and and dispersal of marine benthic and infaunal ele- underground rivers. Second, limno-stygobionts, as ments linked of the with the expansion coastlines and large African inland species represent could have

From Cretaceous deep waterways. the onward the evolved into cave forms from surface water limnic spread of ingolfiellids over the earth might have ancestors. This phenomenon has been extensively taken place at faster into a pace (Fig. 12B) the North studied in recent times by Culver et al. (1995) in

Atlantic and out over the unfolding Indo-Pacific. the case of Gammarus minus. In the case of ingol-

this is fiellids, however, route unlikely because no

epigean relatives are known to exist. Third, the Discussion peracarids may have been interstitial and/or ground-

water forms in origin. This possibility has been little Our justification for interpreting paleomaps and considered. In its favor, however, is the fact that looking for continuous landmasses that have the most may primitive gammaroids (crangonyctids, bogi- bold vast underground bodies of freshwater through diellids) occupy such habitats. The spelaeogripha- long periods of ceans also occur in geological time, is congruent with caves, and Spears (pers. comm.) tbc results of in our analysis which the species believes molecular evidence supports a common

the and mhabiting African Pre-Cambrian shield are pri- origin of spclaeogriphaceans amphipods. In

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 68 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

Cretaceous.

LK Jurrasic;

Late

LJ Jurassic;

early

=

EJ Triassic;

=

Tr. branches.

its

on projected Late -characterstates paleogeographic

with Cladogram

10.

Fig.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 69

P'8- 11. ofthe different of three the of Early parts phylogenetic tree projected on epochs geological history. A, rectangles on map the

Early Triassic the distribution ofthe The the ofthe represent roughly depicted taxa. B, dark grey area on map early Jurassic represents

Eie area adjacent to the Recent mediterranean distribution of the earHest evolved species of the Ingolfiella clade (see Fig. 8). Notice Eds distribution is continuous and restrained to a limited area (maps modified after Scotese, 1997).

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 70 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

Fig. 12. A. The intermediate taxa of the Ingolfiella clade, the fresh and brackish water species of the Caribbean, the Mediterranean,

and South America. The dark the distribution ofthe Late Jurrassic their grey area represents species projected on a map with nowadays

occurrence. B. the most derived taxa ofIngolfiella containing marine species with a scattered distribution virtually worldwide, projected

on of Late a map the Cretaceous (maps modified after Scotese, 1997).

addition, the the most primitive isopods, phrea- and sively or actively invade the insular continental toicids, allied habitats occupy (Wilson, 1998). overviews ground water. Recent (with numerous In contrast, much has been written as how to authors who references to have published on this marine benthic preadapted crustaceans could pas- subject) are presented by Holsinger, 2000; Coineau,

Downloaded from Brill.com10/10/2021 03:55:37AM via free access Contributions to Zoology, 72 (I) - 2003 71

2000; Stock, 1993; Humphreys, 1999; Botosaneanu, gin. Consequently, the occurrence of this family

2001. on the northeastern South American Venezuela-

the of found In summary, by plotting grades taxa Guiana Shield (Magniez, 1981) strengthens the in our a succession of that ancient bodies of freshwater cladogram onto paleographic proposition gave

we can that rise ‘relict’ that be maps perceive shifting patterns sug- to species can found on the frag- gest a possible scenario of ingolfiellidean evolu- ments of Gondwana.

From in tion. an origin the tropic cave and ground We predict that future research on early origins

of that waters Triassic Pangaea, the early progenitors of stygobiont crustaceans concentrates on sam- of the have environments genus Ingolfiella appear to dispersed pling cave deep in the heart lands of eastward along the sub-tropical and temperate shores old cratons in the tropical and temperate climate of the in Jurrasic. Western Thetys Ocean the Early zones, will yield new taxa with ancient origins that

Dispersal in the westward direction along the north- will further elucidate patterns of evolution reach- ern and western coasts of Gondwana happened in ing back to the Triassic or even earlier. the Late As Jurassic. the Atlantic continued to open through the Cretaceous, species of Ingolfiella found themselves isolated on the proto-Atlantic and Car- Acknowledgements ibbean islands. One can envision that some mem- bers of the rode the seafloor down group spreading We are much obligedto Stefan Koeneraann,Sandro Ruffo, Gary

for as the Atlantic deep was created. Afterward further Poore, Jos Notenboom, and Jorgen Berge commenting on the manuscript,'Dr. Traudl Krapp for her indispensable mediative dispersal into the deep Tethys and south Atlantic role, and Angelo Bigontina and Simona Seveso (grant allowed the ingolfiellids to reach the far reaches of ICP94NL3041/43) for translating Italian works on the ecology the Indo-Pacific and southern Ocean in the Ceno- and biogeography of ingolfiellideans. We thank Dirk Platvoet zoic. for his in and active assistance SEM photography original sugges-

tions and Jan Arkel for his Finally, the cladogram we obtained (Fig. 8) sug- van computer imaging. gests, as we pointed out above, that the African and Italian cave species may have given rise to References smaller interstitial freshwater species and these have evolved further via brackish forms into marine Barnard JL, Karaman GS. 1983. Australia as a major evolu- interstitial and A way to test this deep-sea species. tionary centre for Amphipoda (Crustacea). Mem. Austr. Mm. working hypothesis would be to sample for large 18: 45-61. cave Ingolfiellidea in the eastern parts of South Berge J. 2000. A taxonomic revision and cladistic analysis of the America. The of freshwater amphipod family Stegocephalidae (Crustacea: Peraca- presence large cave rida). Dissertation, Univ Tromso, Norway. ingolfiellideans would lend credit to the idea that a Bousficld EL, Shih CT. 1994. The phylctic classification of freshwater continuum underground in Pangea ex- amphipod crustaceans: problems in resolution. Amphipaciflca isted before the break-up of the African and South 1(3): 76-134

American landmasses in the Early Cretaceous. In Bowman TE,Abele LG. 1982. Classification ofthe recent Crus- tacea. In: Bliss DE, ed. The Crustacea, a far in biology of VOI. I. inland cave southwestern Brazil (Gruta do Systematics, the fossil record and biogeography. New York: Lago Azul), a large bogidiellid belonging to a newly Academic Press, 1-319. erected genus, Megagidiella, was found recently Bowman TE, Iliffe TM. 1985. Mictocaris halope, a new unu-

(Koenemann & The size of this sual from Holsinger, 1999). peracaridan crustacean marine caves on Bermuda.

species is exceptional in relation to the widely dis- J. Crust. Biol. 5: 58-73.

tributed Botosaneanu L. 2001. rudimentation and nov- smaller sized bogidiellids. This situation Morphological elties in Cirolanidae Vie is stygobitic (Isopoda, Cymothoidea). more or less comparable to what is seen in ingol- Milieu 51 (1-2): 37-54. i.e., cave far inland that fiellideans, large species Coineau N. 1977. Le genre Pseudingolfiella aux Kerguelen. are rare and found in with ‘relict’ places faunas, as Extrait de CNFRA 42: 287-293.

opposed to the small species from the species-rich Coineau N. 2000. Adaptations to groundwater life. In: Wilkens

interstitial H, Culver DC, Humphreys WF, eds. the world habitats closer to the sea. The strictly Ecosystems of stygobiont 30: subterranean ecosystems. Elsevier, 189-210, freshwater isopod family Stenasellidae has Coleman CO, Myers AA. 2000. New amphipoda from Baltic a distribution that points to a Cretaceous ori- amber. Polsk. Arch. Hydrobiol. 47: 457-464.

Downloaded from Brill.com10/10/2021 03:55:37AM via free access 12 R. Vonk & F.R. Schram - A phylogenetic and biogeographic analysis

Culver DC, Kane TC, Fong DW. 1995. Adaptation and natu- in the gammaridean family Pardaliscidae. Can. J. Zool. 45:

ral selection in caves. Cambridge: Harvard University Press, 347-355,

1-223. Noodt W. 1959. Estudios sobre Chilenos de crustaceos aguas

E. 1977. The model and its subterraneas II. de subterraneas Dahl amphipod functional bearing Nueva Ingolfiella aguas

Zool. 6: 221-228. limnicas de las Lomas Norte de Chile. In- upon systematics and phylogeny. Scripta de Paposo en el

Ebach MC, Humphries CJ. 2002. Cladistic biogeography and vest. Zool. Chilenas 7: 7-16.

the art of discovery. J. Biogeography 29: 427-444, Noodt W. 1965. Interstitielle Amphipoden der konvergenten

Hansen 1903. The fam. of und IIJ. Ingolfiellidae, n., a new type Gattungen Ingolfiella Hansen Pseudingolfiella n. gen.

Amphipoda. J. Linn. Soc. (Zool), London 29: 117-133, plates aus Sudamerika. Crustaceana 9: 17-30.

14-15. Richter S, Scholtz G. 2001. Phylogenetic analysis of the

Ilsiq HU, Mardenbol J, Vail PR. 1987. Chronology offluctuat- Malacostraca (Crustacea), J. Zool. Syst Evol. Research 39:

ing sea levels since the Triassic. Science 235: 1156-1167. I 13-136.

J. 2000. and Ruffo S. 1951. Holsinger Ecological derivation, colonization, Ingolfiella leleupi n. sp. nuovoanfipodotroglobio

speciation. In: Wilkens H, Culver DC, Humphreys WF, eds. del Congo Belga, Rev. Zool. Bot. Afr. 44: 189-209.

Ecosystems of the world 30: subterranean ecosystems. Ruffo S. 1964. Studi sui crostacei anfipodi LVII: Un nuovo

Elsevier: 399-416, delle dell’Africa di Sud ingolfiellide acque sotteranee Ovest.

Humphreys WF. 1999. Relict stygofaunas living in sea salt, Boll. Zool. 31: 1019-1034.

karst and calcrete habitats in arid northwetern australia con- Ruffo S. 1969. Studi sui crostacei anfipodi LX III: Descrizione

tain ancient In: Ponder eds. di many lineages. W, Lunney D, Melaingolfiella mirabilis n. gen. n. sp. (Crustacea,

The other the conservation and inverte- delle 99%, biodiversity of Amphipoda, Metaingolfiellidae fam. nova) acque

brates. Tr. Roy. Zool. Soc. New South Wales, 219-227. sotterranee del Salento nell’Italia meridionale. Mem. Mus.

Just J, Poore G. 1988. Second record ofHirsutridae (Peracarida: Civ. Star. Nat., Verona. 16: 239-260.

Hirsutia from South- Ruffo S. 1970. de la Mictacea): sandersetalia, new species, Considerations a propos systematique et

eastern Australia. J. Crust. Biol 8(3): 483-488. de la biogeographic de Ingolfielles (CrustaceaAmphipoda).

Karaman SL. 1959. Ueber die Ingolfielliden Jugoslawiens, Livre centenaire Emile G. Racovitza, 223-230.

Bioloski Glasnik 12: 63-80 Schram FR. 1981. Late Paleozoic crustacean communities. J.

Koenemann S, Holsinger JR. 1999. Megagidiella azul a new Paleo. 55: 126-137.

Seotese CR. 1997. genus and species of a cavernicolous amphipod crustacean Paleogeographic atlas, Paleomap progress

from Brazil, with remarks its of Univ. (Bogidiellidae) on biogeo- report 90-0497, Department Geology, Texas

graphic and phylogenetic relationships. Proc. Biol. Soc. Wash. Arlington, 1-37.

112; 572-580. Shokita S. 1992 Aquatic animals ofthe Shiokawa salty spring,

Koenemann S, Holsinger JR. 1999. Phylogenetic analysis of Molobu Museum Bulletin, Okinawa 6: 21-44

the amphipod crustacean family Bogidiellidae, s. lat., and Slewing R. 1963. Zur Morphologie dor aberranten Amphi-

revision oftaxa above the species level. Crustaceana 72 (8): podengruppe Ingolfiellidae und zur Bedeutung extremer

781-816. Kleinformen fur die Phylogenie. Zool Anz. 171: 76-91.

Kocncmann S, Vonk R, Schram FR. 1998. Cladistic analysis of Stock JH. 1976. A new member of the crustacean suborder

37 mediterranean from Bonaire, with Bogidiellidae (Amphipoda), including Bogi- Ingolfiellidea a review ofthe entire sub-

diella arista from J. Crust. 18: 383-404. order. Stud. Fauna 50: 56-75. n .sp. Turkey. Biol. Curasao

N. 1955. A de Stock JH. 1993. Some distribution in Leleup propos de Parchaisme et de Tecologie remarkable patterns

VIngolfiella leleupi Ruffo. Notes Biospeologiques 10: 145- stygobiont Amphipoda, J. Nat. Hist. 23: 807-819.

148, Takeuchi I, Ishimaru SI. 1991. Redescription of Capro-

Lincoln RJ. 1979. British Gammaridca. marine Amphipoda: gammarus gurjanovae Kudrjaschov & Vassilenko, 1966

London: British Museum (Natural History). (Crustacea: Amphipoda) from Hokkaido, Japan, with notes

Lowry JK, Poore G. 1989. First ingolfiellids from the south- on the taxonomic status of Caprogammarus. Hydrobiologia

west Pacific (Crustacea: Amphipoda) with a discussion of 223; 283-291.

their Proc. Biol. Soc. Wash. 102 933-946. Vonk R. 1990. systematics. (4): Thalassostygiusexiguus n.g., n.sp., a new marine

Lowry .IK, Springthorpe RT. 2001 onwards. Amphipoda: interstitial melitid (Crustacea, Amphipoda)from Curasao and

Families and Subfamilies.Version 1:1 September 2001. http:/ Klein Bonaire (Netherlands Antilles). Stygologia 5: 43-48. Zcrustacea.net Vonk R, Sanchez E. 1991.A new marine interstitialingolffielld

Maas A,Waloszek D. 2001. Larval development ofEuphausia (Crustacea, Amphipoda, Ingolfiellidea) from Tenerife and

superba Dana, 1852 and a phylogenetic analysis of the Hierro. Hydrobiol. 223: 293-299.

Euphausiacea. Hydrobiologia 448: 143-169. Wilson GDF. 1998. Historical influences on deep-sea isopod

Magnicz GJ. 1981. Biogeographical arid Paleobiogeographical diversity in the Atlantic Ocean. In: Smith CR., Mullineaux in stenasellids problems (Crustacea Isopoda Asellota of un- LS, Levin LA, eds. Deep-Sea Research Part II, TopicalStud-

dergroundwaters). Int. J. Speleol. 11:71-81. ies in Oceanography 45(1-3), “Deep-sea biodiversity: acom- Martin JW, Davis GE. 2001. An updated classification of the pilation ofrecent advances in honor of Robert R. Messier”. recent Crustacea. Science Series of the Nat. Hist. Mus. Los Oxford, UK: Pergamon Press, 279-301.

Angeles County, no, 39: 1-124.

Mills EL. 1967. Deep-sea Amphipoda from the western North

Atlantic Basin I. Ingolfiellidea and an unusual new species Received: 2 September 2002

Downloaded from Brill.com10/10/2021 03:55:37AM via free access