7 \ eu
yes
+

N6h9 Oo
ea Ys Y
Varner

$ »
Fh ahs

4
ELAS f
\Ne

rt a
aw
Serres 3

Hose

bey 3

i=
alan a
awh aks

my
Tet

ah.

Fs,
Hf

fe

Le
iy
im

HARVARD UNIVERSITY
we
Library of the
Museum of

Comparative Zoology

Tas . ; > a i vS ¥ i a

fH |
; 4
ay
7
|
i
f :
r
f ft i ’
pee 4
i lis rae
4 rite Se
Wet
y
‘
t
s4
4
L
‘ ‘
ae
~
{
)
}
i]
1 : ry =
' are
wi! i Pai
Le? J aan)
a bh | i « i 4 .

nt ee
— =
a

7 4
a

—~ oad
4
ae

—
= a
a
- Sieek
y
§

ey

POSTILLA
PEABODY MUSEUM
YALE UNIVERSITY

NUMBER 151 28 JULY 1971

THE FORM AND FUNCTIONS OF THE
AVICULARIA OF BUGULA (PHYLUM
ECTOPROCTA)

KARL W. KAUFMANN

POSTILLA

Published by the Peabody Museum of Natural History, Yale University

Postilla includes results of original research on systematic, evolution-
ary, morphological, and ecological biology, including paleontology.
Syntheses and other theoretical papers based on research are also
welcomed. Postilla is intended primarily for papers by the staff of
the Peabody Museum or on research using material in this Museum.

Editors: Zelda Edelson, Elizabeth G. Weinman, Elise K. Kenney.

Postilla is published at frequent but irregular intervals. Manuscripts,
orders for publications, and all correspondence concerning publications
should be directed to:

Publications Office
Peabody Museum of Natural History
New Haven, Conn., 06520, U.S.A.

Lists of the publications of the Museum are available from the above
office. These include Postilla, Bulletin, Discovery, and special publica-
tions. Postilla and the Bulletin are available in exchange for relevant
publications of other scientific institutions anywhere in the world.

Inquiries regarding back numbers of the discontinued journal, Bulletin
of the Bingham Oceanographic Collection, should be directed to:

Walter J. Johnson, Inc.
111 Fifth Avenue
New York, N.Y. 10003.

THE FORM AND FUNCTIONS OF THE AVICULARIA OF BUGULA
(PHYLUM ECTOPROCTA)*

KARL W. KAUFMANN

Department of Geophysical Sciences, University of Chicago,
Chicago, Illinois 60637

(Received June 10, 1970)
ABSTRACT

The most important function of the bird’s head avicularium of
Bugula simplex and B. stolonifera is that of reducing the tube-build-
ing activities of amphipods. These avicularia would probably also
be effective in reducing predation by animals of such a size and
shape that could be readily seized. That the avicularium is well
adapted to such functions is indicated by:

1) the structurally sound design that maximizes the force that can
be applied for the least amount of material;

2) the arrangement of the musculature that increases the ability to
grasp objects of small diameter;

3) the reaction to mechanical stimuli, namely, nodding and closing
the mandible;

4) the placement of avicularia on the colony in positions where they
can easily seize crawling organisms. Animals effectively seized by
avicularia are those that crawl over the colony and are 0.5 to 4 mm
long with either many appendages or wormlike bodies less than
0.05 mm in diameter. Most animals outside this range of size and
shape, including most potential predators and larvae of fouling
organisms, are inhibited very little by Bugula-type avicularia.

*Publication No. 85 from the Center for Marine and Environmental Studies,
Lehigh University.

POSTILLA 151: 26p. 28: JULY 1971

2 POSTILLA

INTRODUCTION

Avicularia are modified individuals of an ectoproct colony; they
have greatly enlarged opercula and greatly reduced polypides. They
are confined to the Cheilostomata, the most diverse of the three
Recent ectoproct orders, where they occur in a majority of the
families. There is a great diversity of types of avicularia. The shapes
of the mandibles vary from broad and spatulate to narrow and
pointed, and may vary in size by a single order of magnitude (Fig. 1).
Even for similarly shaped mandibles, the form of the entire avi-
cularium may vary drastically (Fig. 2).

Three primary functions have been suggested for avicularia of all
types (Table 1): food gathering, defense and the creation of water
currents. Some of the references have been to specific types of
avicularia, commonly to those found on Bugula, but often all types
were implied to have the same function. Hincks (1880, p. lxxviii)

FIG. 1. Types of mandibles found on avicularia. Note the large range in size
and shape; F, G, H and E, J, O were on the same colonies. [A, E, J, K, M, O,
P: Q)V after Harmer, 1926, pls) xvill, XxXv. XV) XXII xKiexovesexVIIn
XXii, xvi respectively; B, C after Harmer, 1934, pl. xxxix, D, F, G, H, I, L, N,
R, S, T after Harmer, 1957, pls. 1xxx, 1xv, 1xv, 1xv, 1xxiii, 1 xiii, 1xii (fig. 67
on p. 721), 1xv, 1xx respectively.]

AVICULARIA OF BUGULA 3

FIG. 2. Various types of avicularia found on Cheilostomata Ectoprocta.
K shows two vibracula (Vb), polymorphic structures similar to avicularia
(Av). (B, G, H, I, J, K, L, M after Harmer, 1926, pls. xxiii, xxvii, xxviii,
XXVili, XXXii, Xvi, xxix, xxiv respectively; F after Harmer, 1934, pl.
xxxvi; A, C, D, E after Harmer, 1957, pls. 1xii, liv, xiii, 1 respectively.)

and Marcus (1926, p. C 56) dispelled the food-gathering hypothesis
by noting, among other things, that no type of avicularium could
possibly capture the phytoplankton that ectoprocts eat. Harmer
(see Table 1) provided the strongest support for the defense hypo-
thesis. He concluded that avicularia would be especially effective
aganist larvae of fouling organisms, particularly larvae of other
ectoprocts, and against intruders crawling over the colony. Marcus
(1926, p. C 58) noted that the large spacing of avicularia on the col-
ony allowed many small epizoans to become established on the
colony. The third primary function, the creation of water currents,

POSTILLA

(oce 4) 656] “URWAH
(9ST 4) ST6T ‘SIMIOH

(CYS S750) ISS] <=
CCL=6 D606
(98r “d) 9681 ‘owe
(xIxxI ‘d) OST ‘SyourH,
(ZZ 4) ZTL8T ‘seg

Gor Oarss
(gz ‘d) epcgy ‘ysng
(ZL 4) Eps “UYyory

(19 4) OZ6T ‘J9[sseg pue nueD
(9ST ‘d) C16] ‘SIMI9H

(ZZ 4) ZL8T ‘sues

(Cold) 9763
GZ2Oes|———

(SIT 4) 6r8I Asn

(pee 4) Lg ‘uojsuyor

(ZL “d) Epgt ‘UYyory

(61 4) g96] ‘weysey4D

(sg¢ d) 196] ‘IonneH

9¢ 9 6c6l
(p9 ‘19 “d)

OZ6I ‘1oIsseq pue nuedy

(69 4) O06T 9ATRO

(06 4) gggt ‘uorne

(Srz 4) Lest TheATea

($89 “d) Opgt ‘uueWIpION

(07 4) 98LT ‘Jepuejog pur siTq
(9¢ “d) SCL ‘SITA

SLNAYANO
YALVM AO NOILVAAO

SQOYUNV TIAOSIN ASNH4Adad ONIAAHLVS GOOA

VINVTINDIAY AO SdadAL TIV AO SNOILONNYA DHL NO AYNLVYALIT AHL JO AUYVWWNS “[T ATAVL

AVICULARIA OF BUGULA 5

has been described as aiding in respiration, washing away excretory
products or driving the prey toward the tentacles. Hincks (1880,
p. Ixxvii) made the important observation that the large variation in
mandible types indicates that not all avicularia have the same func-
tions. However, the implicit assumption that a particular avicularium
has only one function is common. While this may be true under a
broad definition of the presumed function, such an approach ob-
scures the possibility that the structure is a compromise resulting
from many selective forces acting concurrently. Thus the most fruit-
ful question to ask is not what is the function of an avicularium, but
rather what are the selective forces that have been most important
during the evolution of the avicularium. The answer may then be
phrased in terms of causes rather than results. To ask such a ques-
tion, One must separate what the structure can do, and is thus
available for selection, from what the structure actually does while
interacting with the environment. In a careful analysis of functional
morphology, Bock and Von Wahlert (1965) made this distinction
and explained that although a structure may have many faculties
(the ability to do many things), only a few of these may be important
to the animal. (“Faculty,” as used here, is nearly equivalent to Bock
and Von Wahlert’s use of the term. “Function” is equivalent to their
term, “biological role.”’)

This paper first examines the faculties of the avicularia of Bugula,
and then evaluates the selective forces associated with these avi-
cularia. It amplifies an earlier preliminary report (Kaufmann, 1968).

TERMINOLOGY

The terms, “ventral” and “dorsal,” are used by analogy to the
autozooid where the ventral side is the side of the opening into the
interior, through which the lophophore projects. The various parts
of the avicularium (Figs. 3, 4 and 5) will be referred to as follows:

adductor muscle support — the prominent, bulbous part to which
the adductor muscles are attached

rostrum — a beaklike projection attached to the distal part of the
adductor muscle support

peduncle — the portion proximal to the adductor muscle support

peduncle cushion — the rounded projection from the autozooid

6 POSTILLA

mandible — the appendage attached to the ventral part of the avi-
cularium

cryptocyst — a horseshoe-shaped partition between the rostrum and
the adductor muscle support which continues into the peduncle
and forms a strut across the ventral part of the peduncle

Reference is to the avicularia of both Bugula stolonifera and

B. simplex unless otherwise specified.

ADDUCTOR
MUSCLE
SUPPORT ROSTRUM
CRYPTOCYST
__ SENSORY
BRISTLES
FRONTAL
MEMBRANE
AUTOZOOID
/ MANDIBLE

PEDUNCLE

PEDUNCLE
CUSHION

FIG. 3. Avicularium of Bugula simplex. Only the chitinous and calcitic parts
and the sensory bristles are shown.

AVICULARIA OF BUGULA ie

CRYPTOCYST

DORSAL
SIDE

ADDUCTOR
MUSCLES

PEDUNCLE

TENDON
CUSHION

ABDUCTOR

MUSCLES VENTRAL
CRYPTOCYST SIDE
FRONTAL MEMBRANE
PORE
.05 mm

FIG. 4. Cutawav view through a sagittal section of an avicularium of Bugula
simplex, The sensory bristles and soft parts other than the muscles have
been omitted.

MATERIALS AND METHODS

Avicularia were studied in live specimens and in dried and mounted
material. All of the colonies of Bugula simplex and B. stolonifera
used for observational and experimental work were collected from
the supply department dock in Eel Pond of the Marine Biological
Laboratory at Woods Hole, Massachusetts. The majority of the
colonies came from two tires suspended from floating docks and
covered primarily with ectoprocts, tunicates, barnacles, sponges and
hydroids.

Avicularia were dried or mounted in Lakeside Cement. Mounted
specimens were cleared in a series of alcohols and xylene and
stained in carmine red or methylene blue. Specimens observed with
the scanning electron microscope were dried and coated with 200 A

8 POSTILLA

FIG. 5. Photographs of dried avicularia of Bugula simplex taken with a
scanning electron microscope. A. The mandible is partially open. B. The
mandible is closed and the strut across the peduncle cushion is visible where
the membrane has been pressed around it. C. The mandible is open to its
maximum extent. D. The mandible is completely open and the ventral sur-
face is facing upward. E. The mandible is closed; note the rounded ventral
edge of the rostrum. F. Detail of D; the hinge is best seen in the lower left
corner. The tendon appears as a broad ribbon passing over the support
between the hinges. G. Detail of E; the membrane attached to the curved
support on the mandible has been pressed into the pore in the proximal part
of the mandible. Note the attachment of this membrane to the rim of the
peduncle. H. Detail of E; note the shrunken membrane between the peduncle
and the peduncle cushion. I. Detail of E; note the attachment of the membrane
over the peduncle to the rim of the peduncle. Scale 1 applies to B, C, D, E.
Scale 2 applies to F, G, H, I. A is reduced 20% relative to Scale 1.

AVICULARIA OF BUGULA 9

10 POSTILLA

of gold and palladium. A motion picture of the movement of avi-
cularia was made. Measurements of wall thickness were determined
partially by direct measurement and partially by observing differ-
ences in interference colors under polarized light.

FORM AND STRESS ANALYSIS OF THE AVICULARIUM

The form of an avicularium is a strong one for dealing with stresses
that result from the grasping of an object between the mandible and
the rostrum (Fig. 6A, B). Parts that must withstand compressional
stresses are calcified and shaped so that shear stresses and bending
moments are minimized, thus reducing the amount of skeletal ma-
terial needed, whereas parts that must withstand shear stresses and
bending moments, in addition to compressional stresses, are thicker
and contain even more calcite. Those parts that must withstand
tensional stresses have little calcite and a large proportion of
chitinous matrix.

The wall material consists of a chitinlike matrix with varying
amounts of calcite. Calcite, like most minerals, has high compres-
sional strength and low tensional strength. Its brittleness and ease
of cleavage indicates that it is comparatively weak under bending
moments and shear stresses. It is reasonable to hypothesize that the
organic matrix has a higher tensional strength than calcite, although
I know of no work on the structural properties of the matrix.

The adductor muscle support (Fig. 3) is shaped like a portion of
slightly flattened sphere. The wall is thin (about 0.0024 mm) com-
pared to other parts of the avicularium. The adductor muscles
(Fig. 4) close the mandible. Their insertions are evenly distributed
over the rounded portion of the adductor muscle support. The
muscle fibers join and merge into successively larger bundles toward
the attachment to the tendon.

The nearly spherical shape of the adductor muscle support re-
duces the shear stresses and bending moments that develop (Fig. 6)
when the adductor muscles contract, so that the major remaining
stresses in the wall material are compressional. The shear stresses
and bending moments are at a minimum under the following condi-
tion. The direction of the vector sum of the forces exerted by the
muscles passes through both the center of the arc described by the
adductor muscle support in sagittal section (Fig. 7, M) and the

Gq BENDING MOMENT
| SHEAR STRESS

COMPRESSIONAL STRESS B

TENSIONAL STRESS

-7OoOo Se

mp EXTERNAL FORCE .OSmm

FIG 6. Distribution of stress that would result if the avicularium grasped an
object with the mandible open 50°. Only the components of the forces that
are parallel to the plane of the diagrams are shown. A. Cutaway view of a
sagittal section. Note that several of the symbols are on the part of the
avicularium behind the plane of the section. F; — Force on the adductor
muscle support from the adductor muscles. Fo — Force on the mandible
from the adductor muscles. F; — Reaction forces on the rostrum and man-
dible from the object that is grasped. B. Plane view of the distal part of the
cryptocyst. Fy — Forces on the cryptocyst from the adductor muscles. —
F; — Forces on the fulcrum of the cryptocyst from the mandible.

2 POSTILLA

FIG. 7. Sagittal section of Bugula simplex showing the mandible and adductor
muscles in several positions. In Position A, the mandible is completely closed.
Position B (50° from A) is the point at which the adductor muscle support
can withstand the greatest force from the muscles in an isometric contraction.
Position C (90° from A) is the point at which the tendon (not shown) first
touches the hinge line. Position D (180° from A) is the maximum extent of
opening. Point L is the center of distribution of the adductor muscles. Point M
is the center of the circle describing the arc of the adductor muscle support.
The muscle strands are diagrammatic, but show the most proximal and
distal limits of attachment.

center of distribution of the adductor muscles (Fig. 7, L). This direc-
tion is parallel to the tendon attached to the mandible. Such is the
case when the mandible is opened 50° (Fig. 7, Position B).

The shape of the dorsal margin of the rostrum is similar to that
of the curved compressional stress trajectories developed in a rec-
tangular beam constrained at one end with zero degrees of freedom
and loaded at the other with a force perpendicular to the long axis

AVICULARIA OF BUGULA 13

of the beam. Such stress trajectories are shown by Thompson (1917,
p. 678) in his analysis of the form of bones. Using the same analogy
to a beam Bock (1966) analyzed stresses in birds’ beaks some of
which are shaped much like the rostrum. I conclude then, that the
dorsal part of the rostrum, like the adductor muscle support, sus-
tains minimal shear force and bending moment when an object is
grasped as shown in Figure 6. Thus, the amount of calcite required,
and hence the thickness of the wall, is minimized. The distal part
of the rostrum which is considerably thicker is an exception. But
here, both shear forces and bending moments would be concen-
trated if some live and actively struggling animal were grasped.

The tensional forces that develop when an object is grasped are
concentrated along the ventral edges of the rostrum. As can be seen
in the photographs taken with the scanning electron microscope
(Fig. 5B, G), these edges are rounded rather than knifelike as noted
by Harmer (1909, p. 720). When observed under a polarizing
microscope, however, the isochromes are first-order grey, a fact
indicating that a very small amount of calcitic material is present
and that the ventral edges are mostly chitinlike.

Forces exerted by the adductor muscles to close the mandible
would collapse the adductor muscle support if it were not reinforced
distally and proximally. Reinforcement is provided by the cryptocyst
(Figs. 3, 4). The distal part is a heavily calcified, horseshoe-shaped
structure that separates the rostrum from the remainder of the
avicularium. Two fulcrums about which the mandible rotates are
found on the ventral end of this portion of the cryptocyst. From the
two fulcrums, the cryptocyst extends proximally into the peduncle
and is continuous with a strut across the ventral portion of the
peduncle (Figs. 4, 5B). The thickness and heavy calcification of the
cryptocyst greatly increase its ability to withstand the shear forces
and bending moments that develop when the mandible closes upon
an object (Fig. 6A, B).

The form of the peduncle (Figs. 3, 4) is perhaps the most difficult
part of the avicularium to distinguish with certainty. It consists of a
thickened semicircular wall to which is attached a flexible mem-
brane on the ventral surface. The membrane extends from the
curved ventral and proximal support of the mandible to the peduncle
cushion. It is attached to the edge of the wall of the cryptocyst. The
proximal part of the peduncle at the attachment to the peduncle
cushion is flexible around the entire area of attachment. This can
easily be seen in a motion picture taken of a live avicularium.

14 POSTILLA

The peduncle cushion (Figs. 3, 4) to which the peduncle is at-
tached is a rounded flexible knob that projects from the rim of the
lateral wall of the autozooid just below the orifice. The flexible
nature of the peduncle cushion reduces the shear stresses that would
develop at the point of attachment of the avicularium if the avi-
cularium were pulled or twisted violently. The major stresses that
remain at the point of attachment are tensional. Such pulling and
twisting moments occur when a large organism is caught and
struggles violently. If the avicularium were attached directly to the
autozooid without the cushioning effect of the peduncle cushion,
then a force parallel to the longitudinal axis of the autozooid would
set up shear stresses at the relatively small area of attachment in-
stead of transferring them to the much larger perimeter of the
peduncle cushion.

The mandible (Figs. 3, 4), unlike the rest of the avicularium, is
lightly mineralized, but still rigid. The proximal part is supported by
a bow-shaped structure covered by a membrane with a small pore in
the center. A flat sheet extending from the strut across the hinge
line to the place of attachment of the tendon, further reinforces the
distal half of the mandible. On the distal end is a short spike that
inserts into the rostrum when the mandible is closed. The part of
the U-shaped channel that extends from the spike to the attachment
of the tendon is thickened along the ventral portion, thus increasing
the amount of bending moment and shear stress it can withstand.
Just proximal to the spike, the sides of the U-shaped channel form a
rounded notch. The spike can prevent an object from slipping out
past the end of the mandible as it closes; it can also puncture objects,
but such objects would have to be small enough to fit between the
tip of the rostrum and the tip of the spike.

The mandible is opened by means of two sets of abductor muscles
(Fig. 4). One set depresses the frontal membrane of the peduncle,
causing the mandible to rotate on its hinge, as shown by Silén (1950,
p. 355, fig. 4) for Bugula flabellata. This set of muscles consists of
7-10 strands that extend from the thickened part of the peduncle
wall to the frontal membrane. Some strands converge to a small
region halfway between the peduncle cushion and the mandible. The
other set, which opens the mandible to its maximum extent, consists
of a pair of muscle bundles, each with about five strands. The
bundles are inserted on the adductor muscle support on either side
of, and distal to, the adductor muscle and extend to the frontal
membrane. I could not determine whether they were attached to the

AVICULARIA OF BUGULA 15

frontal membrane, as shown by Silén (1950) or to the bow-shaped
support of the mandible, as shown by Calvet (1900, p. 50, fig. 9) for
B. sabatieri [Calvet’s B. sabatieri is synonymous with B. simplex
(Ryland, 1960, p. 97)]. The position of the avicularium with respect
to the autozooid determines the maximal opening (about 180°) of
the mandible. When the avicularium nods back and forth, the man-
dible opens and closes slightly as it brushes against the surface of
the autozooid.

Sensory bristles project outward from a point just above the hinge
line (Fig. 3). When the mandible is completely open, 50-100
bristles extend as far as an imaginary line drawn from the tip of the
rostrum to the tip of the mandible. Stimulation of the bristles causes
the mandible to snap shut. Often whatever touches the sensory
bristles is caught between the mandible and the rostrum, and is
lodged in the rounded notch just proximal to the spike of the man-
dible. If such an object remains still, it is released after a few sec-
onds, but if the object is continually agitating, the grasp will not be
released.

FORM OF THE COLONY

The branches of colonies of Bugula simplex (Fig. 8A) are 3-5
zooids wide and arranged in fanlike clusters of about five branches
per cluster. Bunches of these clusters are, in turn, arranged in
conical groups that may consist of several concentric layers. The
ventral surface of each zooid from which the polypide projects faces
inward. The overall appearance of the colony is that of a tightly
packed ball of thick branches, as high as 3 cm and several centi-
meters in diameter. Prominent holes appear between the branches.

Colonies of Bugula stolonifera (Fig. 8B) have the same rounded
appearance as B. simplex, but the branches are biserial rather than
multiserial and are not arranged in conical groups. Consequently,
the individual zooids are more evenly dispersed over the region oc-
cupied by the colony and there are no large “‘holes” in the colony,
as exist in B. simplex.

Avicularia are attached to the lateral margins of each branch in
such a way that their nodding motion carries the tip of the rostrum
into the space between branches. There is a space of about 0.5 mm
between avicularia on branches of both species.

16 POSTILLA

FIG. 8. Branches from colonies of Bugula stolonifera (A) and B. simplex (B).
Note the presence of avicularia (Av) on every zooid of the biserial branch
of B. stolonifera, but only on the outer zooids of B. simplex. (A) from Maturo,
1966, p. 572, fig. 12: (B) from Ryland, 1960, p. 92, fig. 13.

THE FACULTY OF GRASPING

The faculty of grasping can best be assessed if the diameter of an
object that can be held securely can be determined. The ultimate
force with which the mandible can close upon an object placed
between it and the rostrum depends upon the size and position of
the object and the moment that the muscles exert through the ten-
don about the hinge.

The length of the moment arm as a function of the angular open-
ing of the mandible (Fig. 9) may be calculated by making the fol-

AVICULARIA OF BUGULA 17

lowing assumptions. The first assumption is that the resultant of
the force of the adductor muscles acting on the mandible will
always be along a line passing through the center of distribution of
the adductor muscles (Fig. 7, Point L). Observations of live avi-
cularia show that this is nearly true for all positions of the mandible.
The second assumption is that the muscles all converge at one point,
where they are attached to the end of the tendon. The region of
convergence is small enough to make this assumption reasonable.

The moment arm is at its maximum when the mandible is closed
(Fig. 9). It decreases rapidly to half when the mandible is opened
to 50° and to less than 1/20 of the maximum when the mandible
opens 90°. Between 90° and 180° opening, the tendon wraps around
the hinge line and the moment arms becomes just as long as the dis-
tance between the axis of rotation and the slightly bowed strut across
the hinge line.

The tension developed in a muscle during an isometric contrac-
tion is not the same at all extensions of the muscle. Mostly vertebrate
muscles have been studied; for these there is one muscle length, the
normal length of the muscle in the body, at which the maximum

1.00
”

Ss = .80
Ca
<Gi>

2 AGO
2a
a:

Sa .40
=a
a

.20

O56 90° 180°
CLOSED OPEN

ANGULAR OPENING OF MANDIBLE

FIG. 9. Graph of the length of the moment arm between the hinge line and
the tendon as a function of mandible opening. Values are expressed as a
fraction of the maximum moment that exists when the mandible is completely
closed.

18 POSTILLA

tension can be developed. At greater or lesser lengths, the tension
is less (Prosser and Brown, 1961, p. 433). From a purely mor-
phological analysis, it is difficult to determine the length of maximum
tension for the adductor muscles. Because some of the muscles vary
considerably in length as the mandible closes, it cannot be assumed
that the tension remains constant. The muscles that are inserted most
distally in the adductor muscle support must contract almost 50%
to close the mandible, and those near the center of distribution of
the muscles (Fig. 7, Point L) must contract about 35%. The muscles
that are most proximal appear to lengthen slightly while the man-
dible rotates from 90° open to 50° open and then contract about
10% in closing the mandible from there.

However, in spite of the lack of knowledge about the changes in
muscle tension relative to length, a more than twentyfold increase
in the moment arm as the mandible closes makes it reasonable to
assume that the force that can be exerted upon an object with the
mandible partially closed is substantially greater than the force that
can be exerted when the mandible is open 90° or more. When the
mandible is opened more than 90°, the avicularium cannot grasp
objects securely because the mandible and rostrum are not suffi-
ciently opposed in this position.

When the mandible is opened to 50°, the moment arm is half the
maximum and an object 0.05 mm in diameter could fit into the
space between the rostrum and the mandible. When it opens to 90°,
the object size that can be grasped increases to only 0.1 mm while
the moment arm is reduced to its minimum. The moment arm, then,
gives the avicularium the faculty of grasping most securely objects
which are at least 0.1 mm and preferably less than 0.05 mm in
diameter.

THE FACULTY OF NODDING

The nodding motion of the avicularium consists of oscillation back
and forth in an arc of nearly 180° in the plane of a sagittal section.
It increases the effective area over which the sensory bristles can
come in contact with an object. If the colony is undisturbed, the
avicularia nod infrequently; Bugula simplex is less active than
B. stolonifera in this respect.

Stimulation of the sensory bristles of a single avicularium of

AVICULARIA OF BUGULA 19

B. stolonifera causes it to nod forward and snap shut. By nodding
forward when an object touches the sensory bristles, the avicularium
prevents the object from slipping past the distal end of the mandible
before it closes completely. If nothing is caught, the avicularium
continues nodding vigorously about once a second. Prodding the
avicularium with a needle without touching the sensory bristles in-
creases the nodding rate less than actually touching them. Stimula-
tion of the tentacles of the individual to which the avicularium is
attached causes no increase in nodding rate, but sticking a needle
into the autozooid or removing the branch from the water briefly
causes all avicularia on the branch to start nodding vigorously.
Moving the branch around under water increases the nodding
slightly.

Rey (1927) and Forbes (1933) found many chemical stimuli that
affect the nodding rate, but the relevance of these to the natural
environment is not clear.

Silén (1950, p. 350-358), working with Bugula flabellata and
four other species of Bugula, hypothesized that the adductor muscles
were responsible for nodding. These muscles pull on the membrane
covering the peduncle, causing the avicularium to nod dorsally
while using the lateral sides of the peduncle as a pivot. The resiliency
of the dorsal part of the peduncle near the peduncle cushion causes
the avicularium to nod ventrally when the muscles relax. This model
required that the membrane on the ventral side of the peduncle be
attached below the dorsal edge of the peduncle.

Avicularia of Bugula stolonifera and B. simplex do not fit Silén’s
model. In B. stolonifera, the membrane remains limp throughout the
nodding motion and becomes taut only when the mandible snaps
shut. Thus, the membrane cannot take part in the nodding. In addi-
tion, rather than the lateral sides of the peduncle acting as a pivot,
the avicularia are separated from the peduncle cushion by a thin
membrane around the periphery of attachment (Figs. 3, 4 and 5H).
The ventral edge of the peduncle is compressed inward at a point
near the peduncle cushion as the avicularium nods ventrally. This
deformation was seen clearly in a motion picture of live avicularia
of B. stolonifera; however, the orientation of the avicularia on
B. simplex prevented a clear view the peduncle of this species being
deformed. Photographs of both species taken with a scanning elec-
tron microscope indicate that the membrane is attached to the
ventral edge of the peduncle (Fig. 5B, E), although the photographs
are difficult to interpret because of shrinkage and deformation which

20 POSTILLA

may occur when the specimens are dried. These changes could cause
the membrane to be pulled inward and appear attached below the
rim of the peduncle in some (Fig. 5A) but not all (Fig. 5 I) photo-
graphs.

As far as I could determine, Calvet’s (1900, p. 53) explanation
of the nodding motion in Bugula simplex is correct. This explana-
tion implies that the peduncle deforms with both ventral and dorsal
movements of the avicularium, instead of only with the dorsal move-
ment as implied by Silén. However, I could not find the extensor
and flexor muscles described by Calvet in the peduncle.

INTERACTION OF AVICULARIA WITH THEIR ENVIRONMENT

Many have observed the strength of avicularia. They can grab a
needle “so firmly the branch might be shaken” (Darwin, 1839,
p. 259), or capture various small organisms (Harmer, 1896, p. 485),
but no one has evaluated the strength of avicularia in terms of the
frequency of capture and ability to hold organisms commonly found
on the colony.

A wide variety of organisms inhabits colonies of ectoprocts col-
lected in the field. Among the more common were nematodes,
gammarid amphipods, and copepods. The gammarid amphipods
(Corophium insidiosum and Jassa falcata) range in size from 0.5 mm
to 4.5 mm and build tubes composed partly of ectoproct fecal ma-
terial in the lower parts of the colonies where the polypides have
degenerated into brown bodies. In late summer, when most of the
polypides have degenerated, amphipods may have tubes covering
the colony. I have counted as many as 100 amphipods inhabiting
tubes or crawling about on a colony of B. simplex 2 cm high. Less
common animals include triclad flatworms less than 1 mm long,
pycnogonids and marine mites. Ciliates, hydroids, and more rarely
tunicates, sponges and several species of ectoprocts, are also attached
to colonies.

The interaction between larger organisms and avicularia is quite
remarkable. I have observed gammarid amphipods (Corophium
insidiosum and Jassa falcata) 4 mm long, not counting the antennae,
caught repeatedly by avicularia no longer than 0.2 mm. The few
amphipods larger than 4 mm escape from avicularia readily, and
those that remain in their tubes are not captured. Most often amphi-

AVICULARIA OF BUGULA 21

pods are caught by one of the hairs on their appendages and escape
after two or three minutes of struggle. Occasionally an amphipod is
caught by one of the terminal segments of the antennae and held
as long as 37 hours. The larger amphipods (1-4 mm) wandering
about on the colony, spend more than half their time trying to free
themselves from the avicularia. By far the most frequently captured
animals are amphipods. When present, nematodes larger than 1 mm
and smaller than 3 mm long are also readily captured. A pycnogonid,
10 mm across the legs, is held securely by an avicularium. Smaller
nematodes, marine mites and flatworms are rarely captured.

Harmer (1909, p. 720) claimed that avicularia discourage the
settlement of larvae by crushing or puncturing them. Another possi-
bility is that larvae are prevented from settling because either the
avicularia or the lophophores brush them away as they approach the
colony. To test this hypothesis and to determine whether anything
unusual happens in the presence of larvae, I observed the interaction
between ectoproct larvae and Bugula-type avicularia.

Sixteen ectoproct larvae were placed in a shallow fingerbowl con-
taining a brush of Bugula simplex oriented with its lophophores ex-
tended upward. Many attempts were made to use the phototactic
response of the larvae to maneuver them toward the branch, but
very few actually made contact with anything but the lophophores.
Often, a larva was drawn into a lophophore and held for a short
time (as long as a full second) by curling the tips of its tentacles
around the larva. It would then be ejected as is normally the case
for a particle too large to be ingested. No larvae were caught or
hindered by the mandible of the avicularium. Thus, the lophophores
are a far more effective instrument for discouraging the settlement
of larvae than the avicularia.

DISCUSSION OF THE FACULTIES AND THEIR FUNCTIONS

The best developed faculty of the Bugula-type avicularium is that of
grasping objects about 0.05 mm or less in diameter. The forms of
the individual parts, particularly the adductor muscle support, the
cryptocyst, and the large moment arm for objects less than 0.05 mm
in diameter, are structurally very sound for applying great force
through the mandible. The intermittent nodding motion, the reac-
tion to tactile stimuli by snapping the mandible shut and the spike on
the mandible aid in capturing animals. The flexibility of the peduncle

22 POSTILLA

cushion prevents the avicularia from being twisted off by animals
that are caught.

The limitations of the faculty of grasping animals can be de-
termined if the placement of the avicularia on the colony is con-
sidered. First, the manner in which avicularia move in a plane
parallel to the surface of the branches means that animals crawling
on the surface are more easily caught, and animals swimming above
the surface less easily caught, than if the avicularia moved in a plane
perpendicular to the surface. Swaying into the holes in the branches
of colonies of B. simplex increases the probability that the avi-
cularium will come in contact with an animal crawling about on the
colony, because such an animal must invariably crawl through these
holes. Second, the spacing of about 0.5 mm between avicularia
along the branches permits small animals to crawl about with a low
probability of encountering an avicularium, whereas larger animals,
particularly those that are longer than the spacing between avicularia,
will be caught more frequently. Factors other than size and mode
of movement which determine the frequency that animals come in
contact with avicularia are the number of appendages suitable for
being grasped, and the amount of commotion an organism makes
in moving over the colony. Amphipods are usually grasped by one
of the terminal segments of an appendage. Nematodes smaller than
1 mm are captured infrequently, presumably because of their lack
of appendages, while those larger are captured regularly because
their violent twisting movement causes their bodies to cover a large
area in their progress over a colony.

The ability of the larger amphipods to escape can be attributed
both to their strength and to the increased diameter of the terminal
segments on their appendages. The upper size limit for the capture
of many organisms of the same general shape as amphipods is
probably about the same. Small annelids with many small hairs and
spindly pycnogonids, although larger than 4 mm, could probably be
captured.

In comparison, the faculty of creating water currents and of
brushing larvae or sediment off the colony is very poorly developed
in Bugula-type avicularia. The compact form of the avicularium
and the pointed tip of the mandible could not move water nearly as
efficiently as a broad spatulate mandible (Fig. 1F), nor is the
sporadic nodding motion capable of creating a continuous flow. In
species of Bugula, at least, lophophores create a much stronger cur-
rent flow through ciliary action. For brushing larvae and sediment

AVICULARIA OF BUGULA 23

away, a long narrow mandible or a vibraculum (Fig. 2) would be
much more effective than a compact avicularium covering a limited
part of the colony.

Because the faculty of grasping is so well developed, there must
have been strong selective pressures for it. The most obvious func-
tion for the faculty of grasping is that of defense. Some animals may
harm the colony but are either too small (ciliates), too large (fish and
most molluscs), or are not the right size to be captured often enough
to reduce their threat to the colony and be affected by the avi-
cularia. Small snails have very few appendages and do not move
actively about the colony.

Smaller annelids, nematodes and crustaceans may crawl about
on the colony, but just this alone does not appear to damage the
colonies. Pycnogonids (Barnes, 1963, p. 377) and tanaid amphipods
(Smith, 1906, p. 335) are predators of ectoprocts, but I observed
few pycnogonids and no tanaids on the colonies I studied. Avicularia
could be valuable controls of these predators where they are more
common.

Larvae deserve special mention because many workers have sug-
gested that avicularia of most species aid in discouraging their settle-
ment. The avicularia studied here cannot defend against larvae
effectively. Most common fouling organisms have larvae that are
less than 1 mm long and would as often find a place to settle between
avicularia or on the dorsal surface of the colony as come in contact
with an avicularium. More important, most larvae of fouling or-
ganisms have few appendages suitable for being grasped, and are
too large to be grasped by the body. The larvae of most ectoprocts,
for example, are 0.15—0.2 mm in diameter with only short cilia
protruding from their bodies. For an avicularium to grasp the entire
body, it would have to open wider than 90° (Fig. 5, Position C) and
the moment arm would then be reduced to its minimum. Nodding
would certainly prevent larvae from settling within the small range
of movement of the avicularia, but because the nodding movement
is closely associated with the faculty of grasping, its separate use
for the function of brushing away larvae would appear incidental
to its function of increasing the probability that the sensory bristles
will come in contact with an animal.

Only the amphipods, Corophium insidiosum and Jassa falcata,
do extensive damage through building their tubes in the colonies,
thus preventing the lower parts of the colonies from regenerating
new polypides and producing larvae. Avicularia do reduce the

24 POSTILLA

amount of tube building by capturing and holding amphipods that
leave their tubes. Holding the amphipods for varying periods of time
does not cause the amphipods to leave, but it does cut, by at least
half, the time that they have available for building more tubes.

CONCLUSION

Grasping and retaining small organisms crawling over the colony
are the most highly-developed faculties of the avicularia of Bugula
simplex and B. stolonifera. These faculties reduce the tube-building
activities of amphipods and can reduce the harm done by other
organisms, such as predators of ectoprocts, provided shape, size and
activity on the colony allow them to be caught frequently. Avicularia
on other species of Bugula appear to have an equally well-developed
faculty of grasping and probably have similar functions.

ACKNOWLEDGMENTS

The encouragement and valuable suggestions of Professor Thomas
J. M. Schopf of the University of Chicago, at every stage of this
paper are greatly appreciated. The study was supported by Thomas
J. M. Schopf through National Science Foundation Grant GB 7325.
Most of the work was carried out at the Marine Biological Labora-
tory, Woods Hole, Massachusetts. The Rogick Collection of books
and reprints at the library of the Marine Biological Laboratory was
especially helpful. Dr. Frank Maturo, University of Florida, identi-
fied B. stolonifera, and Mr. Allen Michael, Dalhousie University,
Halifax, Nova Scotia, identified the amphipods. Dr. R. D. Allen,
State University of New York at Albany, allowed the use of his
time and facilities to make a motion picture of avicularia. Mr. Akira
Kabaya of JOELCO, Inc., donated the use of their scanning electron
microscope model JSM-2 and assisted in taking the photographs.

AVICULARIA OF BUGULA 25

LITERATURE CITED

Barnes, Robert D. 1963. Invertebrate Zoology. W. B. Saunders Co., Phila-
delphia. 632 p.

Bock, W. J. 1966. An approach to the functional analysis of bill shape. Auk
83: 10-51.

Bock, W. J. and G. von Wahlert. 1965. Adaptation and the form-function
complex. Evolution 19: 269-299.

Busk, George. 1849. Observations on the sheppard’s purse coralline of Ellis
(Notamia bursaria Fleming). Trans. Microscop. Soc. London 2: 110-112;
Pl. xxv.

1854a. Catalog of the marine Polyzoa in the collection of the British

Museum. Cheilostomata Part II. p. iv—vii, 55-120; Pls. 69-124.

1854b. Remarks on the structure and function of the avicularian
and vibracular organs of the Polyzoa; and on their value as diagnostic
characters in the classification of these creatures. Trans. Microscop. Soc.
(N:S.), 2: 26-33; PI. ii.

Calvet, Louis. 1900. Contributions a l’histoire naturelle des Bryozaires
Ectoproctes marins. Théses présentées a la Fac. des Sci. de Paris.
Montpellier. 488 p., 13 pls.

Canu, F. and R. S. Bassler. 1920. North American early Tertiary Bryozoa.
U.S. Nat. Mus. Bull.106. 2 vol. 879 p., 162 pls.

1929. Bryozoa of the Phillippine Region. U.S. Nat. Mus. Bull. 100.
Vol. 9. 685 p.

Cheetham, Alan. 1968. Morphology and systematics of the Bryozoan Genus
Metrarabdotos. Smithsonian Misc. Collections 153(1) Pub. 4733, 121 p.,
18 pls.

Dalyell, J. G. 1847. Rare and remarkable animals of Scotland. John van
Voorst, London. Vol. 1. 268 p., 53 pls.

Darwin, Charles. 1839. Journal of researches into the geology and natural
history of the various countries visited by H.M.S. Beagle. Henry Col-
burn, Great Marlborough St., London. 615 p.

Ellis, John. 1751. An essay towards a natural history of the corallines and
other marine productions of the like kind, commonly found on the
coasts of Great Britain and Ireland. Privately published, London. 103 p.,
39 pls.

Ellis, J. and Daniel Solander. 1786. The natural history of many curious and
uncommon zoophytes. Benjamin White and Son, London. 206 p., 63 pls.

Forbes, Alexander. 1933. Conditions affecting the response of the avicularia
of Bugula. Biol. Bull. 65: 469-479.

Gautier, Yves Victor. 1961. Recherches écologiques sur les Bryozoaires
Chilostomes en Méditerranée occidentale. Théses présentées a la Fac.
des Sci. de l'Université D’Aix-Marseille. Aix-En-Provence. 431 p.

Harmer, Sidney F. 1896. Polyzoa. Vol. II, p. 465-491. Jn S. F. Harmer and
A. E. Shipley [eds.] The Cambridge Natural History. Macmillan & Co.,
Ltd., London.

1909. Address to the zoological section. Section D. Report Brit. Ass.

Advance. Sci., 78th meeting. Dublin, 1908. John Murray, London,

p. 715-731.

1926. The Polyzoa of the Siboga Expedition Part II, Cheilostomata

Anasca. p. i—vill, 181—501; Pls. xii—xxxiv.

1931. Recent work on the polyzoa. Presidential address to the

Linnean Society of London. Linn. Soc. Proc. 143: 113-168.

26 POSTILLA

1934. The Polyzoa of the Siboga Expedition Part III, Cheilostomata

Ascophora I, Family Reteporidae. p. i—vi, 502-640; Pls. xxv—xli.

1957. The Polyzoa of the Siboga Expedition Part IV. Cheilostamata
Ascophora II. p. i-ix, 641-1147; Pls. xlii—Ixxiv.

Herwig, E. 1915. Die avicularien von Bugula flabellata. Archiv fir Natur-
geschichte. Embrik Strand, Berlin. Part A. 7: 156-159.

Hyman, Libbie H. 1959. The Invertebrates. McGraw Hill Book Co., New
York. Vol. V. 783 p.

Hincks, Thomas. 1880. A history of the British marine Polyzoa. John van
Voorst, London. Vol. I. 601 p.

Johnston, George. 1847. A history of the British zoophytes. 2nd ed. John
van Voorst, London. Vol. I. 488 p.

Jullien, J. 1888. Bryozoaires. Mission Scientifique du Cap Horn 1882-83.
Vol. VI, Zoologie, Part 3. 92 p., 15 pls.

Kaufmann, Karl W. 1968. The biological role of Bugula-type avicularia.
Proc. of the First Int. Conf. on Bryozoa, 1968. Atti della Societa
Italiana di Scienze Naturali e del Museo Civico di Storia Naturale di
Milano 108: 54-60.

Krohn, August. 1843. Uber die vogelkopfahnlichen und ihren verwandten
Organe bei den Bryozoen. Jn Froriep und Froriep [eds.] Neue Notizen
aus dem Gebiete der Natur- und Heilkunde. 25: 70-72.

Marcus, Ernst. 1926. Bryozoen. Vol. 4, Part 7C, p. C1—C100, In G. Grimpe
and E. Wagler [eds.! Die Tierwelt der Nord- und Ostsee; Akademische
Verlagsgesellschaft, Leipzig.

Maturo, Jr., Frank J. S. 1966. Bryozoa of the southeast coast of the United
States: Bugulidae and Beaniidae (Cheilostomata: Anasca). Bull. of
Marine Sci. 16: 556-583.

Nordmann, Alex von. 1840. Recherches microscopiques sur la Cellularia
avicularia. Vol. I, p. 679-707. In Voyage dans la Russie méridionale
et la Crimée, exécuté en 1837 sous la direction de M. Anatole de Demi-
doff. E. Bourdin et Cie., Paris.

Prosser, C. Ladd and Frank. A. Brown, Jr. 1961. Comparative animal
physiology, 2nd ed. W. B. Saunders Co., Philadelphia. 688 p.

Rey, P. 1927. Action de divers agents sur les mouvements des aviculaires
de Bugula. Bull. Soc. Zool. France 52: 367-379.

Ryland, J. S. 1960. The British Species of Bugula (Polyzoa). Proc. Zool. Soc.
London 134: 65-105, 3 pls.

Sars, George Ossian. 1872. On some remarkable forms of animal life from
the great deeps off the Norwegian Coast Part I. University Programme,
Christiania. 82 p., 6 pls.

Silén, Lars. 1950. On the mobility of entire zooids in Bryozoa. Acta Zool.
31: 349-386.

Smith, Geoffrey. 1906. High and low dimorphism. With an account of cer-
tain Tanaidiae of the Bay of Naples. Mittheilungen aus der zool. Stat.
zu Neapel. 17: 312-340.

Thompson, D’Arcy Wentworth. 1917. On growth and form. Cambridge
University Press, Cambridge. 793 p.

i Acme

Bookbinding Co., Inc.
300 Summer Street

* Boston, Mass. 02210

: > ENE eI

3 2044 066 305

——

Sb Cea uty

€:¢
rh

is
i eepit
neg Ge

(yin

(ents
Whe f Mee
: ae Gl

2 BOs
Cr a
Sieve ta ve

ou

ie
wath

iris
sats
fr.
iit
ve (‘ate \

et
Heren di
er ii

ee ;

yj
errs
miftaes vt

f
ie phate ;
JRE

. &

%

wait gt ae ak

| had
eS Ory Rew ara
CDA ed)

NAP DSI E &
Ska es aes
Mean eked ‘

.
ie

Reng

eae
HVS eet
WY de gk E