0 OESTRACT Measurements were made of the force required to break the barnacle Halanus glandula while the animal was attached to tour different rocky substrata. Force was exerted on the barnacle normal to the substrate surface and the dislodgement force was recorded. In general, force was not proportional to barnacle size and barnacles were stronger om stronger rocks. The results are discussed in the context of the environmental hydrodynamic forces encountered and the population genetics of B. glandula. INTFODUCTION The larvae of Beglandula must undergo two metamorphoses before becoming a sessile adult. The barnacle during these two stages is pelagic and free-swimming, but the second stage larvae, referred to as the cyprid, must settle permanently before the second metamorphosis occurs. Ihis settlement takes place on a wide variety of substances. whose physical properties, such as surface strengt, can vary greatly. Strength is the amount of force per area that must be applied to cause the pieces uf the material to break. he barnacle has three types of body plates: basal, shell, and opercular. The basal plate, which forms the base of the barnacle's truncated cone-like shape, cannt be cemented to the shell plates, which comprise the side walls, if the barnacle is to expand its diameter during growth. Consequently, 'fixation fibres' hold these two plate types tagether. Cement is exuded beneath the calcareous basal plate to anchor the barnacle to the substratum. Os barnacles cam enjoy a fairly lang lifetime, ane would expect that they would attempt to maximise the strength of this anchoring system. The cement af the adult balanoid barnacles has beem studied with regard to its strength of adhesion to slate, steel, and rubber surfaces (Carderelli,18; Yule ? Walker,1784) and the work of Yule and Walker (1984) provides a stress value of 1.7 X 10 Fa for barnacle cement before cohesive tailure occurs. Yet even knowing the breaking stress value (another term for strength) for the cement, one still does not know how biplogically relevant this value is. Although the cement may be able to withstand stresses of um to 170,000 Fa, other components, such as the substratum to which the barnacle is attached, the shell plates, or the fixation fibres may not. The present study was designed to see what part of the entire barnacle-rocky substratum complex would fail first, and at what stress level. The experiments were carried out on four different rock types to test for any effect of the substratum. MATERIALS AND METHODS B. plandula were studied at three locations near Monterey, California comprising four different rocky substrata: (1) The north side of Whale Feak at Garrapata State Fark, located about S miles south of Carmel, with two types of granite, (2) Stillwater Cove, in Carmel Bay next to the Febble Beach golf course, with a type of sandstone, (I) Natural Bridges State Fark in north Santa Cruz, CO, with Monterey shale. For each test, a straightened paper clip was used to remove the barnacle's body from the shell and a "O-Tip" was then used to absorb as much of the remaining fluid as possible. Ofter the inside of the barnacle was allowed to dry for a few minutes, "Five Minute" epoky was applied to the inside of the barnacle shell. A pre-bent paper clip, with a loop at one end and a right angle bend at the other, was then placed into the epoxied shell and held normal to the substratum, until the epoxy hardemed enough to maintain the paper clip in this position (see Figure 1). The length and width of the barnacle's basal area were then measured to the nearest millimeter with a ruler. Length here is defined as the longest line in the area of contact between substrate and barnacle; width is the largest dimension perpendicular to length (see Figure 2). For the tests where the substrate broke, the length and width were measured again after the specimen was dislodged. These measurements were compared to the original measurements obtained and if any discrepancies existed, the latter set of values was used. The bazal area was assumed to e best approximated by an ellipse and therefore was calculated using the equation: Orea - T X (length/2) X (width/2). Ofter the epoxy had hardened, a recording spring scale (zimilar to that used by Jones and Demetropoulos, 1968) was used to determine the force required to remove either the barnacle, the substratum to which it was attached, the paper clip, or some combination of rock and barnacle. A carabimer attached both to the spring scale and through the loop of the modified paper clip made certain that the sample would not be lost in the process of removal (see Figure 3). Every attempt was made to administer a steady pull, although this was accomplished by hand. Once dislodged, the sample was examined to determine what part of it had brokem, and the force required for removal was noted. Simple linear regressions were run on the tests in order to determine if a relationship existed between the basal area of the barnacle and the force needed to dislodge the specimen. These were performed for each group in the study, a group being defined by the rock type the barnacle was attached t and the location of the break. RESULTS 123 tests were performed. For 35 of these, the paper clips were pulled out, while the rack and the barnacle remained attached; these tests will not be considered further. In only pne case was breaking force correlated wtih area: that for the first type of granite when the barnacle broke. The relationship was as follows: Force required to break barnacle (Newtons)- 272424.à Fa X Basal area (m) + 2.87 M (r -.4401, 29 D.F.). For the forty-two tests where the barnacle broke, the mean breaking stress values differed significantly among rock types, with the exception of those barnacles on the sandstone and shale (see Figure 4 and Table 1). Twenty-one of the 12 tests broke the rock before the barnacle or paper clip dislodged. The granite broke just once (this was the first type of granite), while the sandstone and shale broke a total of twenty times. The mean surtace breaking stress for the Monterey shale found in Santa Cruz was approximately three times greater (161,338.89 Pa) than that for the Stillwater Cove sandstone (49,350.00 Pa). Ihe one case of the granite breaking was for a feldspar crystal in the quartz and mica matrix and required 295,000 Pa to dislodge it. Otherwise, all that can be said for the two types of granite is that at least JO2,077.42 Pa for the first type and 523,500 Fa for the second must be applied before they break. These values come from the mean breaking stresses of the barnacles on the granite substrata, since granite did not break before the barnacles did. Additionally, twenty-five tests broke some combination of barnacle and sandstone or shale (see Tables T and 3). For both groups, analysis of variance demonstrated mo significant differences between mean breaking stress values. DISCUSSION As larger barnacles receive larger wave forces tham smaller ones due to a greater exposed area, one would expect that they would have to be attached with a greater temacity in order to survive. Unexpectedly though, the results obtained here do nat support this line of reasoming. Only barnacles attached toone of the four types ofrock demonstrated a significant relationship between force and area. When a B. glandula suffered a break, the basal plate usually remained fixed to the substratum, while the shell plates broke free from it. In order for this to occur, the fixation fibres holding the two types of plates together must be broken. The fracture between the base and the shell was not always "clean"; often part of one or several of the shell plates remained attached to the base and consequently, the fracture split the shell plate itself. In this latter case, the fixation fibres probably were not broken. This variability of the severing fracture's location within B. glandula could contribute to not finding a force versus area relationship, as different components of the barnacle's structural system will most likely posses varying breaking stresses. Several experimental factors should be considered whem analyzing this study. First, forces normal to the rocky substratum were the only ones considered here. Because waves strike the barnacle from many directions, forces in other directions, such as parallel to the rock, need to be also studied. Second, it is possible that the entire basal area of B. glandula is not the correct biological parameter to attempt to relate to force. For instance, a significant relationship has been demonstrated between the force required for removal of Balanus balangides and the cement area present between it's cuticular basal plate and the slate surface to which it is attached (Yule and Walker, 1784), although the actual cement area is different from the basal area. Thus, relating basal area to force might be contributing to the lack of a relationship between force and area noted here. Twenty-five tests resulted in breaking both the barnacle and the rock on which it sat. Here also the variability in location of break probably contributed to the negative finding. Ihe lack of a force versus area relationship for the twenty-ane tests where the rocky substratum broke beneath the barnacle is not too surprising, considering the highly variable treatment rocks are subjected to in the intertidal rone. A variety of organisms affect the surface characteristics of the sutstratum either by physical actions, such as burrowing, or chemical means, such as secreting mucus on the rock which contains a wide variety of chemical compounds. These compounds may alter the integrity of the bonds within the rock which comprise its strength. Fhysical forces from the sun and the waves also affect different portions of the rock to varying degrees. Since these factors weaken the rock's breaking strees in different amounts and their location of attack is more-or-less random, one must expect to find a variety of values for the rock's breaking stress when testing at different sites on the rock. Only a small number of tests were performed for many of the groups, which defined by break location and the rock type on which the barnacle sat. As mentioned above, some of these group sizes would most likely not increase, no matter how many tests were conducted. For other groups though, additional tests would prove relatively easy to obtain. If tests belonging to these latter groups were to be obtained in larger numbers, a more confident and knowledgable analysis of possible trends could be completed. The cyprid larvae of Balanus balanoides have been shown to be able to detect and respond to different characteristics of the surface on which they are exploring as a possible permanent home site (Yule and Crisp, 1983). In the conteat of these studies, the results shown in Figure 4 suggest that at some time in the life cycle, B. glandula is able to determine the strength of the rock on which it sits. If it realizes that it resides on a rocky substratum whose surtace strength is low, it apparently diverts energy away from producing a stronger cement, attachment fibers or shell. This raises the possibility that the barnacles on weak substrata invest energy into the more rapid growth o¬ the reproductive system. This speculation assumes that the populations present at the three locations studied are genetically similar. If they are not, then this scenario may turn put to be simply an example of differences between populations. Unfortunately, not enough is known about the genetic diversity of B. glandula on the coast of California to determine whether this study deals with one or more populations, although this field of research is an active one (personal communication, Lani West). Therefore, any conclusion concerning the results contained in Figure 4 will have to wait until more conclusive studies have been conducted on the species' genetic diversity. Ihe mean sur face strength of the sandstone found at Stillwater Cove is roughly one-third of that of the north Santa Cruz shale and this shale value is approximately half that of the minimum strength for the weaker of the two types of granite (see Tables I and 2). Although the values for the sandstone and shale are not significantly different, the wide mange encompassing all the rock surface strengths demonstrates that the argument in the previous paragraph is plausible, as they are direct evidence for the existence of substrata with differing surface strengths. ACKNOWLEDGEMENT: First and for most, I would like to thank Mark Denny for the time he spent helping me perform and think about my study. I would also like to thank Doug Stomer for assistance with statistics and Lani West for the information concerning the population genetics of B.glandula. Judy Thompson was helpful in a wide variety of ways; additionally, the rest of the staff here at Hopkins have heen both friendly and helpful. Thanks to Allen and the library staff for minimizing my wandering time before actually finding what I wanted. Finally, thanks to my fellow 175H students for providing thought-provoking comments and simply for their friendship. HEFERENCES Carderelli, N. F., 1968. Barnacle Cement as a Dental Restorative. 49 pp. Maryland: National Institute of Dental Hesearch. IFublication no. 151.J Jones, W. E. and A. Demetropoulos, 1968. Exposure to wave action: Measurements of an important ecological parameter on rocky shores on Onglesey. J. Exp. Mar. Biol. Ecol. 2: 46-63. Yule, A. B. and D. J. Crisp, 1783. Adhesion of cypris larvae of the barnacle, Balanus Balanoides, to clean and arthropodin treated surfaces. J. Mar. Bicl, Oss. U. K. 63: 251-271. Yule, A. B. and G. Walker, 1784. The adhesion of the barnacle, Balanus Balanoides, to slate surfaces. J. Mar. Biol. Ass. U. K. 64: 147-156. Table 1 BREAKING STRESS VALUES OF BALANUS GLANDULA ROCK TIPE MINIMUN MAXIMON MEAN 50 STRESS (N/mXm) 302077.42 51500 133488.76 721000 5500.00 162000 890000 302511.16 89630.7 166433 64300 32000 57200 191475.00 142904.20 315000 ROCK TYPE Table 2 BREAKING STRESS VALUES OF SUBSTRATA MEAN MININUM MAXIMUM S.D. STRESS (N/mXm) 295000.00 49350.00 47200 51500 3040.56 161338.89 149396.76 46800 630000 Table 3 BREAKING STRESS VALUES OF A COMBINATION OF BALANUS GLANDULA AND SUBSTRATUN ROCK TYPE MEAN MINIMUM MAXINUN S.0. STRESS (N/ mXm) 42900 146325.00 138840.13 343000 136938.10 146206.05 43600 734000 Figure 1: Diagram of a prepared specimen. Note that the barnacle cement layer is exaggerated in size for purposes of clarity. Figure 2: Diagram of the top view of the barnacle's basal area, illustrating the measurement conventions used in this study. Figure 3: Diagram of the apparatus used to measure the breaking stress value. Figure 4: Graph of rock type versus the mean strength of E. glandula. The middle symbol of the three points for each rock type is the mean value, while the upper and lower points represent the range of the 957 confidence interyals. Analysis of variance and the GTT method were used to obtain the values shown. Figure 1 EPOXV BASAL PLATE ROCK —PAPER CLIE BARNACLE SHELL BARNACLE CEMENT ZLAVER Figure 2 TOP VIEW BASAL AREA LENG WIDTH Figure 3 STEEL WIRE PAPER CLIF 1 „RECORDING SPRING SCALE ZCARABINER BARNACLE — — Figure 4 MEAN BREAKING STRESS VALUES OF BALANUS GLANDULA ON FOUR DIFFERENT ROCK TYPES 600000 500000 0000 300000 0000 100000-