Gross Morphology of Tilapian Intestine Cells Tissues Organs 2000;166:294–303 295IntroductionFish are a remarkably diverse taxonomic group, con-sisting of approximately 20,000 living species [Lagler etal., 1975]. Utilization of a wide spectrum of niches pro-viding a tremendous variety of food items is reflected inextensive variation in the complexity and disposition ofthe fishes’ alimentary canal. Indeed, the structural diver-sity of the intestine among fish has been described asgreater than that in any other vertebrate group [Stevens,1988]. Earlier reviews addressing these features includeKapoor et al. , Stroband and Debts , Sis et al., Clark and Whitcomb , and Ezeasor and Sto-koe [1980, 1981]. Much of the variation in alimentarytract morphology is expressed in the intestine, whichranges from a simple tube with little or no coiling in spe-cies such as salmon and trout, through moderate to com-plex degrees of looping and/or coiling in species such ascarp and ocean sunfish [Suyehiro, 1941; Harder, 1975].Tilapian fish are members of the large and diversefamily Cichlidae. Though gross intestinal morphology hasbeen examined in several members of this family [Ya-maoka, 1985, Reinthal, 1989], tilapia are a large group ofcichlid fish in which intestinal morphology has not beenentirely described. Tilapia are currently utilized in severalmanners which raise interest in their anatomy beyond theself-evident academic interest: tilapian fish figure promi-nently worldwide in aquaculture [Ackefors et al., 1994],and are also proving notable as a laboratory animal [Hartet al., 1997; Augusto et al., 1996]. Thus, a detailed charac-terization of the normal gross morphological features ofthe intestinal tract of tilapia will be of interest to a widesegment of the scientific community. This study beginssuch characterization by describing the complex grossanatomy of the intestinal tract in mature individuals ofthe Nile tilapia, Oreochromis niloticus.MethodsDetermination of in situ RelationsTwo 23-cm (both male) and eight 12- to 13-cm fish (4 males and4 females) were used to determine the in situ emplacement of theintestinal tract. Following a 12-hour fast, each fish was anesthetizedusing tricane methanesulfonate (MS-222, Sigma Chemical Co., St.Louis, MO, USA) and killed by cervical separation. The right and leftbody walls covering the body cavity were removed and the organsexamined in situ from each side (fig. 1).Identification of Intestinal RegionsFourteen adult fish (8 male and 6 female) ranging in size from12.5 to 18.5 cm were examined. The digestive visceral mass and asso-ciated organs (esophagus, stomach, spleen, liver, gall bladder andintestine) were removed en masse by cutting the esophagus as farcranially as possible and cutting the intestine caudally where it joinedthe body wall. The tight mesenteric attachments within the visceralmass permitted its easy removal from the body cavity without dis-turbing topographical relations among its various portions. Theentire visceral mass was immediately immersed in 10% neutral buff-Fig. 1. In situ position of the visceral masswithin the body cavity, left view. From thisaspect, the liver (L), the proximal (PH) anddistal (DH) limbs of the intestinal HL, thePCpL of the PMC (Cp 1), and the intestinalTS (T) are visible. Note the fat lying betweenthe intestinal coils. The gall bladder (GB)and gonads (G) are also visible. White ar-rows indicate direction of ingesta flowingtoward the center of the first major coil. Bar= 2 cm.
12.5296 Cells Tissues Organs 2000;166:294–303 Smith/Smith/Tengjaroenkul/Lawrenceered formalin. To promote fixation of the delicate intestinal wall ade-quate to permit extensive manipulation with minimal breakage, themass was fixed at room temperature for a minimum of 3 days. Thefixative was changed on the 2nd day.After fixation, topographical relations were first documentedwith the digestive visceral mass and associated organs intact. Subse-quently, each digestive tract was dissected to determine features andrelations of the inner coils. Beginning at the stomach, the intestinewas progressively dissected free from its mesenteric attachments toother surrounding structures. As regions of the tract were identified,length measurements of each were made.Table 1. Sequential organization of the fivemain regions and their subdivisions of theintestinal tract of O. niloticusHepatic loopProximal limbDistal limbSpiral intestineProximal major coilProximal centripetal loopProximal centrifugal loopGastric loopProximal limbDistal limbDistal major coilDistal centripetal loopDistal centrifugal loopTerminal segmentStatistical AnalysisThe means, standard deviations, and ranges of the fish length andtotal intestinal length (TIL) were determined. The lengths of the indi-vidual intestinal segments as well as the percentage of the TIL con-tributed to by each segment were also determined. Linear regressionsof the TIL, individual segment length, and percentage of segments onfish length were calculated. Intestinal segment lengths were evaluatedfor gender differences by analysis of variance with inclusion of thecontinuous variable fish length as a covariate in the model. Evalua-tions were performed using the SAS (Sas System for Windows,Release 6.12; SAS Institute, Cary, NC) for all analyses. Specifically,the UNIVARIATE, REG, CORR (Pearson and Kendall’s tau-b) andGLM procedures of this system were used in the evaluations.ResultsTable 1 presents the sequential organization of the fivemain intestinal regions and their subdivisions in O. niloti-cus. Table 2 shows the number, sex, and size of individualfish examined, as well as the lengths of individual intesti-nal segments and the percentage of TIL for each segment.Table 3 displays the mean, standard deviation, and mini-mum/maximum of fish length and individual intestinalsegment lengths, as well as those values for percentage ofTIL provided by individual intestinal segments.The general form of the intestine was an unadornedtube, with pyloric ceca entirely absent. Surface features ofthe various gut regions were essentially similar among allregions identified. Though a faint tendency was presentTable 2. Intestinal tract in O. niloticus:length of individual segments, total length,and individual segments’ percentage ofthe TILFishlengthcmFishsexHLcmPMCcmGLcmDMCcmTScmTILcmm 65 (21) 109 (35) 26 (8) 87 (28) 21 (7) 30812.5 f 72 (22) 115 (35) 25 (8) 88 (27) 28 (9) 32813:0 f 52 (17) 131 (41) 18 (6) 87 (28) 27 (9) 31513.0 m 51 (17) 109 (36) 21 (7) 95 (32) 25 (8) 30113.0 f 52 (18) 121 (41) 24 (8) 80 (27) 19 (6) 29613.2 f 59 (18) 125 (37) 25 (7) 99 (30) 26 (8) 33414.8 f 74 (78) 129 (12) 58 (14) 141 (34) 20 (5) 40215.0 f 66 (22) 103 (33) 35 (12) 77 (26) 12 (4) 31415.5 m 49 (16) 107 (35) 50 (16) 96 (33) 12 (4) 29415.7 m 57 (17) 124 (37) 41 (11) 96 (29) 14 (4) 33216.0 m 65 (20) 98 (31) 45 (14) 89 (28) 21 (7) 31816.0 m 74 (16) 154 (34) 52 (12) 139 (31) 32 (7) 45116.0 m 70 (17) 135 (33) 54 (13) 125 (31) 24 (6) 40818.5 m 69 (17) 138 (34) 70 (17) 106 (26) 25 (6) 408Numbers in parentheses report the percentage of a given intestinal segment of the TIL inindividual fish.
Fish length, cmGross Morphology of Tilapian Intestine Cells Tissues Organs 2000;166:294–303 297Fig. 2. In situ position of the visceral masswithin the body cavity, right view. From thisaspect, the centripetal limb of the PMC (Cp1), two centrifugal limbs of the PMC (Cf 1),the medial surface of the HL curving aroundthe caudal border of the liver, and the intesti-nal TS (T) are visible. The cranial extremityof the liver (L) is also visible. Note the fatlying between the intestinal coils. White ar-rows indicate direction of ingesta flowingtoward the center of a coil, while blackarrows indicate flow away from a center. Bar= 2 cm.for more distal regions of the gut to be slightly smaller indiameter than more cranial regions, this feature was bothinconsistent and also greatly affected by the presence orabsence of fixed peristaltic waves along the gut wall. Shortsegments excised from any one region of the gut wereessentially indistinguishable from any other. Thus, thesole reliable means of precise identification of a particulargut region depended upon the ability to visualize its topo-graphical relations to other gut regions and/or otherorgans.Five principal gross regions of the intestinal tract wereidentified. Progressing caudally from the stomach, thesewere designated as: (1) hepatic loop (HL); (2) proximalmajor coil (PMC); (3) gastric loop (GL); (4) distal majorcoil (DMC), and (5) terminal segment (TS) (fig. 1–6). Thefirst four of these regions each possessed a reversal flexureand thus could be divided into proximal and distal limbs.Only the TS was straight and undivided.The intestine exited the cranial region of the stomachat an acute angle, adjacent to the esophagus and to the leftof the midline (fig. 4, 6). The intestine immediatelyturned dorsally and entered the HL. The proximal limb ofthe HL coursed caudally closely following the dorsalhepatic border, reflected cranially around the caudal edgeof the liver, and continued cranially as the distal limb ofthe HL (fig. 1, 3, 4, 6). At a level approximately even withthe stomach, the distal limb of the HL turned mediallyand entered the spiral region of the intestine (fig. 1, 3,Table 3. Fish length, intestinal segment length, and percentage ofTIL contributed to be individual intestinal segments: means, stan-dard deviations, and minima/maximaVariable Mean SD Minimum Maximum14.6 1.8 12.5 18.5HLcm 62.5 9.0 49.0 74.0% of total 18.3 2.0 16.4 22.0PMCcm 121.3 15.5 98.0 154.0% of total 35.5 3.1 30.8 41.6GLcm 38.9 16.3 18.0 70.0% of total 11.1 3.8 5.7 17.2DMCcm 100.4 20.5 77.0 141.0% of total 29.1 2.8 24.5 35.1TScm 21.9 6.1 12.0 32.0% of total 6.4 1.6 3.8 8.6TIL, cm 343.5 51.0 294.0 451.0
298 Cells Tissues Organs 2000;166:294–303 Smith/Smith/Tengjaroenkul/LawrenceFig. 3. Visceral mass removed from thebody cavity, superficial surface, left view.The proximal (PH) and distal (DH) hepaticloops are visible in their course around theliver (L). The PCpL (Cp 1) and PCPL (Cf 1)of the PMC of the spiral intestinal region areplain. Note the fat between the intestinalloops. The TS (T) of the intestine and the gallbladder (GB) are visible. White arrows indi-cate direction of ingesta flowing toward thecenter of a coil, while black arrows indicateflow away from a center. Bar = 5 cm.Fig. 4. Visceral mass removed from thebody cavity, superficial surface, right view.The dorsal exit at an acute angle of the proxi-mal HL (PH) from the stomach (S), the GL,three PCpL of the PMC (Cp 1), and a por-tion of the intestinal TS (T) are visible. Theliver (L) and spleen (Sp) are also demon-strated. Note how the proximal HL followsthe dorsal border of the liver. Also note thatthe GL emerges from the left side, ap-proaches the stomach, and then reverses di-rection to regain the left side and reenter thespiral intestinal region. The esophagus (E)and gall bladder (GB) are also visible. Whitearrows indicate direction of ingesta flowingtoward the center of a coil, while blackarrows indicate flow away from a center. Bar= 4 cm.table 1). The spiral intestinal region was disposed as atruncated cone lying partially to the left of the midline andoriented dorsoventrally, with its base dorsally and apexventrally (fig. 1–4). The spiral intestine comprised thePMC and DMC, each with its own centripetal and centri-fugal limbs (table 1, fig. 4, 6).The PMC was positioned superficially (fig. 1–4). Thiscoil began with the proximal centripetal limb (PCpL) thatdescended on the outside of the cone, requiring three tofour turns to gain its apex (fig. 1–4). The PCpL thenpassed through a U-turn and, now as the proximal centri-fugal limb (PCfL), ascended the cone deep (internal) tothe previous segment (fig. 2–4). Glimpses of this (deeper)PCfL were occasionally visible in the intact mass betweenloops of the proximal centripetal coil (fig. 2). The PCfLalso required three to four turns to traverse the cone.On gaining the dorsal border of the cone, the PCfLentered the GL (fig. 4, 6). The proximal limb of the GL34
Gross Morphology of Tilapian Intestine Cells Tissues Organs 2000;166:294–303 299Fig. 5. Visceral mass removed from thebody cavity, deep structures, left view. TheHL and PMC (PCpL and PCfL) have beenremoved. The liver has been slightly dis-placed dorsally to demonstrate the GL. Theentrance from the proximal centrifugal coil(Cf 1) into the GL has also been displaced tosimplify its demonstration. Note again howthe gastric loop approaches the stomach (S)and then reverses direction. In this view, thecontinuation of the distal limb of the GLinto the initial portion of the DCpL (Cp 2, ofthe DMC) is evident. Similarly, the origina-tion of the DCfL of the DMC (Cf 2) from thereversal flexure, and its continuation up thespiral intestinal region is also demonstrated.The gall bladder (GB) and liver (L) are visi-ble. White arrows indicate direction of inges-ta flowing toward the center of a coil, whileblack arrows indicate flow away from a cen-ter. Bar = 4 cm.Fig. 6. Visceral mass removed from thebody cavity, deep structures, right view. TheHL and PMC (PCpL and PCfL) have beenremoved. The exit of the proximal HL (PH)from the stomach (S) at an acute angle is visi-ble, as is the initial course HL along the dor-sal border of the liver (L). The GL is shownoverlying the structures from the oppositeside. The disposition of coils in this areamakes following direction of the intestinedifficult. The distal end of the PCfL of thePMC (Cf 1) is shown entering the GL(ghosted in on the opposite side of the stom-ach). The DCpL of the DMC (Cp 2) contin-ues from the GL back to the left side, to enterthe spiral intestinal region. The DCfL (Cf 2)of the DMC are also visible, as is the intesti-nal TS (T). E = Esophagus; GB = gall blad-der. White arrows indicate direction of in-gesta flowing toward the center of a coil,while black arrows indicate flow away from acenter; open arrows indicate flow in intesti-nal segments lying deep to the stomach andGL in this view. Bar = 4 cm.56turned cranially and progressed until it nearly touched thestomach. At this point it passed through a U-turn and par-alleled its own proximal limb to reenter the dorsal portionof the spiral region by becoming continuous with theDMC.Both limbs of the DMC lay entirely internal to the limbsof the PMC. The DMC began with the distal centripetallimb (DCpL) (fig. 5, 6), which descended just deep to thePCfL. Being more internal, the turns of the DCpL weretighter than those of the preceding limbs. The DCpL typi-cally required three turns to traverse the cone. The DCpLpassed through a U-turn at the cone’s apex, and continuedas the distal centrifugal limb (DCfL) (fig. 5, 6). This rever-sal flexure of the DMC was intimately nested with thereversal flexure of the PMC. From this reversal flexure, theDCfL ascended to the dorsal surface of the cone by passingthrough three turns internal to the DCpL. Being the mostinternal of the four spiral limbs, the turns of the DCfL were
300 Cells Tissues Organs 2000;166:294–303 Smith/Smith/Tengjaroenkul/Lawrencethe tightest of all regions of the spiral intestine. From here,the intestine turned caudally, straightened, and graduallyextended ventrally en route to the anus as the TS (fig. 1–6).The cranial region of the TS paralleled and lay in directcontact with the dorsal border of the hepatic limb. A fold ofmesentery attached the TS to the liver.Statistical analysis showed that male fish in this study(n = 8, mean length = 15.4 cm, SD = 1.9 cm) tended to belonger than female fish (n = 6, mean length = 13.6 cm,SD = 1.0 cm, p = 0.06). However, after adjustment for fishlength, gender did not affect either length of intestinal seg-ments or percentage of individual segments of TIL. Noeffect of fish length on the length of most intestinal seg-ments (HL, PMC, DMC and TS) was observed. However,TIL [r2 = 0.36, p = 0.02; TIL = 95.2 + 17.0 ! fish length(FL)] and length of the GL (r2 = 0.83, p ! 0.01; GL = 82.1+ 8.3 FL) increased with fish length. The percentage ofTIL contributed to by the PMC (r2 = 0.28, p = 0.052) hada tendency to decrease, and the TS (r 2 = 0.76, p ! 0.01)absolutely increased.DiscussionTilapia are similar to most other species of fish in lack-ing gross surface distinction among various intestinalregions. Only the tight mesenteric attachments and subse-quent constant topographical relation of one area toanother permitted immediate identification of the var-ious regions in the intact or partially dissected intestinalmass.The absence of intestinal ceca in the Nile tilapia mayrelate to their herbivorous/omnivorous diet. Though in-testinal ceca are present among various species of fishincluding both carnivores and herbivores, ceca tend to bebetter developed in carnivorous than herbivorous fish,and best developed in carnivorous fish with short intesti-nal length [deGroot 1971]. The role of intestinal ceca haslong been debated, with numerous functions having beensuggested including absorption, fermentation, storage,and breeding sites for gut microbiota [Saddler and Ashley,1960; Reifel and Travill 1978, 1979]. However, cecalfunction has subsequently been more clearly demon-strated as absorptive and similar to that of the cranial por-tion of the intestine. Buddington and Diamond provided convincing evidence that the ceca serve anabsorptive function similar to that of the cranial intestineby demonstrating an essentially similar uptake of variousnutrients in the ceca as in the cranial intestine. More qual-itatively, Hossain and Dutta supported these findings byfirst  showing that the ceca develop from mucosalfolds of the intestine rather than the stomach (indicatingtheir derivation from an intestinal region dealing withabsorption), and later  by demonstrating that theexpansive capacity of intestinal ceca is minimal (whichrules out a prominent role in food storage). Given thatintestinal ceca likely function mainly to increase theabsorptive area of the gut, the possession of a long intes-tine by O. niloticus would obviate the need for ceca.Indeed, when Buddington and Diamond  pointedout gut absorptive area of fish increases with either length-ening the gut or adding ceca, they noted ‘tilapia’ (withoutmentioning genus name) as an example of the former.Apparently the length of the continuous tubular portion ofthe Nile tilapia’s gut accommodated by the intricate coil-ing of the intestine provides sufficient surface area toenable the fish to derive adequate nutrition from its herbi-vorous/omnivorous diet.The percentage of total length of a given intestinal seg-ment displayed fair consistency among adult O. niloticusof varying size (table 2). The GL was the only segmentthat departed notably from this generalization. In smallerfish (12–13 cm), the GL averaged about 7% of the TIL,while in larger fish (15–23 cm) the GL was significantlylonger, averaging 13.5% of the TIL. In addition to propor-tionate change, statistical evaluation showed the increasein absolute length to be significant. The longer GL in larg-er fish contributed to the other length change observed,which was significantly longer TIL in larger fish. Thoughother features of length demonstrated no significant dif-ferences, a tendency was observed for the proportionatelength of the PMC and TS to decrease. The increase in theabsolute and percentage length of the GL likely contrib-uted to this effect.Among fish (as also among most other vertebrate spe-cies), greater intestinal length and more complex disposi-tion of the gut tube is generally characteristic of herbi-vores and onmivores, while shorter intestinal length andsimpler disposition usually typifies carnivorous species[Al-Hussani, 1949; deGroot, 1971; Harder, 1975; Reifeland Travill, 1979; Zihler, 1982; Geevarghese, 1983; Rib-ble and Smith, 1983; Korovina et al., 1991; Menin andMimura, 1992; Kramer and Bryant, 1995]. Considerableoverlap in TIL exists among these trophic groups. Harder, summarizing the work of Jacobshagen ,reported that intestinal length relative to body length incarnivorous fish generally ranges between 0.2 and 2.5times body length, in omnivorous fish, between 0.6 and8.0 times body length, and in herbivorous fish, between0.8 and 15.0 times body length. The Nile tilapia thus falls
Gross Morphology of Tilapian Intestine Cells Tissues Organs 2000;166:294–303 301within the range characteristic of herbivorous fish, albeitnearer the shorter end. The relatively short length of thetilapian intestine as compared to other herbivorous fishlikely relates to the tilapian’s well-known ability to readilyadapt to different foodstuffs: though adult tilapia arelargely herbivorous, they are quite opportunistic and free-ly make insects and crustaceans a significant part oftheir diet [Philippart and Ruwet, 1982; Trewavas, 1983;Wheeler, 1985].Patterns of intestinal looping are constant within, butvary tremendously among, the thousands of fish species[Mohsin, 1962; Harder, 1975; Kapoor et al., 1975; Reifeland Travill, 1979; Zihler, 1982; Reinthal, 1989]. Coilingof the intestinal tube in most fish is relatively simple, typi-cally exemplified as a simple sigmoid curve, a screw-typespiral, or a flat disc with all coils occupying the same plane[Suyehiro, 1941; Harder, 1975]. Even in fish character-ized by a spiral colon, such as the bitterling (Rhodeus seri-ceus amarus), the spiral typically consists of a single cen-tripetal limb passing into and a single centripetal limbextending out of the spiral, with no loops that leave andthen return to the spiral [Harder, 1975]. Certain otherspecies such as the mullet (Mugil cephalus) and the gold-fish (Carassius auratus) [Harder, 1975] and the algal feed-ing cichlids Cyathopharynx lucifer and Ophthalmotilapianasuta [Yamaoka, 1985] show a more complex loopingpattern, but such instances are not as common.Zihler  described the gross morphology and con-figuration of the digestive tracts of 71 species of cichlidfish (not including Oreochromis). Allowing for variationrelated to general body form and feed preferences, he con-sidered the general morphological characteristics of thedigestive tract of the family Cichlidae as synapomorphic.In many respects, O. niloticus conforms with the patterndescribed by Zihler as characteristic of cichlid fish. Suchfeatures include the extensible blind pouch of the stom-ach, the exit of the intestine from the left side of the stom-ach, and the first intestinal loop lying on the left side ofthe body. Zihler  described several forms of intesti-nal looping among cichlids, ranging from short and simpleto lengthy and complex, again related to diet. Comparedto these patterns, the looping pattern of O. niloticus ismost similar to the coiling pattern identified by Zihler astype H and exemplified by the African cichlid, Sarothero-don mossambicus. Both O. niloticus and S. mossambicuspossess a cone-shaped mass of intestines and a relativelyshort gastric loop (referred to by Zihler as the ‘flap-backloop’). However, S. mossambicus does not demonstratethe doubled form of the spiral colon (i.e., does not possessboth the proximal and distal major coils).Among the species described by Zihler , cichlidfish were characterized as possessing a form of intestinalloop arrangement he defined as convoluting. In the con-voluting form, the intestine leads once into a spiral andonce out. The intestinal morphology of the Nile tilapia [aswell as certain other cichlids; Reinthal, 1989], however,departs from this pattern and instead more closely followsa pattern described by Zihler  as coiling. Zihler described the coiling form of intestinal dispositionas leading twice into a spiral region and twice out, with thereversal flexures of the two sets of loops being closelynested into each other.Complex patterns of intestinal coiling have indeedbeen previously characterized in cichlid fish. In additionto characterizing some of these patterns, Yamaoka also demonstrated that their ontogeny can be used todetermine phylogenetic relationships among some spe-cies. Reinthal  described the intestinal morphologyof six genera and 16 species of cichlid fish (not includingOreochromis). His study confirmed that longer intestinallength is associated with a diet high in plant material (aswell as in species feeding mainly on detritus), but alsodescribed an intestinal coiling pattern similar to that ofO. niloticus. Most species he described possessed an intes-tine that passed into, then out of a spiral, through a looprelated to the stomach (termed by him the ‘haplochrom-ine’ loop, i.e. the ‘flap-back’ loop of Zihler and the gastricloop of this study), then back into and out of the spiralregion again. This pattern of coiling was largely similaramong most species he examined, with the main varia-tions he observed lying in the number of loops in eachcoil. Thus, O. niloticus fits the general pattern describedby Reinthal as characteristic of cichlids. However, O. nilo-ticus is unique in the manner of disposition of the spiralintestine’s coils. The quadruple-looped spiral intestine ofthe Nile tilapia, with each successive loop lying internal tothe preceding loop, was not included in Reinthal’s de-scriptions. Thus, though adhering to certain generalizedfeatures of the cichlid family, O. niloticus also nonethelesspresents certain striking unique features as well.The coiling of the intestinal tract in the Nile tilapiaplainly achieves the self-evident advantage in accommo-dating into the body cavity an intestinal tract many timeslonger than the cavity itself. However, an intricate patternof looping and coiling is unnecessary for this goal, whichcan be achieved as simply and uncomplicatedly as suspen-sion of unfixed lengthened intestinal loops from a singlelong mesentery, as in the jejunum of many other verte-brates. Thus, the complex nature of the Nile tilapia’sintestinal coiling pattern suggests that some goal other
302 Cells Tissues Organs 2000;166:294–303 Smith/Smith/Tengjaroenkul/Lawrencethan this most apparent feature may also be served. Onepossible basis could be that, beyond simply adding lengthto the gut, the topographical disposition of the loops insome manner favors absorption of nutrients from the gut.Characteristics that would slow passage of ingesta throughthe intestine and/or increase the efficiency of nutrientabsorption from the intestinal wall would contribute tothis effect. The Nile tilapian intestine passes through fourreversal flexures. Three of these flexures (between the twoparts of the PMC, GL and DMC) are extremely acute,180° turns. Flow rate would necessarily be slowed in orderto complete these turns. Also, the course followed by theingesta through the three acute reversal flexures as well asthrough the immediately succeeding segment of gut re-quires ascending against the pull of gravity, which couldalso tend to slow the rate of ingesta passage. As an addi-tional if less likely possibility, the exact paralleling of theproximal and distal limbs of the PMC and DMC (as wellas of the GL) together with the opposite direction of inges-ta flow within each of these respective limbs suggest somechance of some form of counter- or cross-current mecha-nism in these regions. Unpublished results of studiesinvolving distribution of various digestive and absorptiveenzymes through the length of the Nile tilapian gut showthat digestive and absorptive enzymes are indeed plenti-ful in the PMC and GL, and also present in the DMC,though they are lacking in the TS. Investigations as to thenature of the blood supply to these regions may provideinsight as to whether such mechanisms are indeed at workin these areas.In summary, the intestinal tract of O. niloticus wascharacterized by a series of loops set in a constant andintricate pattern that is both unique among species de-scribed to date, and also one of the more complex patternsreported in fish. The intestine departed the stomach, fol-lowed the elongate borders of the liver, and entered thespiral region of the intestine. The spiral intestine con-sisted of two paired major coils (proximal and distal),each of which comprised a centripetal and centrifugalloop. The short gastric loop (‘flap-back’ or haplochromineloop) was interposed between the two major coils. The ter-minal segment of the intestine departed the spiral regionand followed a straight course to the anus.Possession of an intestine of a length greatly exceedingthat of the body cavity as well as the disposition of thatelongated gut into loops or coils of some sort, as possessedby O. niloticus, are characteristic of adult herbivorous fishin general. With a TIL approximating 2.5 times the totalbody length, the Nile tilapia falls nearer the shorter end ofthe range of 0.8–15 times body length characterizing mostherbivorous fish. This characteristic may reflect the fish’sadaptability in diet, which can be readily modified fromits typical diet of phytoplankton to include animal food inthe form of crustaceans.AcknowledgmentsThe authors would like to thank the Commercial Fish and Shell-fish Technology (CFAST) program of Virginia Polytechnic Instituteand State University for providing the funding for this project, andMs. Sandy Brown for maintenance of the fish in the Aquatic Medi-cine Laboratory of the Virginia-Maryland Regional College of Veteri-nary Medicine.ReferencesAl-Hussani, A.H. (1949) On the functional mor-phology of the alimentary tract of some fishesin relation to differences in their feeding habits:Anatomy and histology. Q J Microsc Sci 90:109–139.Ackefors, H., J.V. Huner, M. Konikoff (1994) In-troduction to the General Principles of Aqua-culture. New York, Food Products Press,p 122.Augusto, J., B.J. Smith, S.A. Smith, J. Robertson,R. Reimschuessel (1996) Gentamicin-inducednephrotoxicity and nephroneogenesis in Oreo-chromis nilotica, a tilapian fish. Dis Aquat Or-ganisms 26: 49–58.Buddington, R.K., J.M. Diamond (1986) Pyloricceca of fish: A ‘new’ absorptive organ. Am JPhysiol 252: 665–676.Clark, A.J., D.M. Whitcomb (1980) A study of thehistology and morphology of the digestive tractof the common eel (Anguilla anguilla). J FishBiol 16: 59–170.deGroot, S.J. (1971) On the interrelationship be-tween morphology of the alimentary tract,food, and feeding habits of flatfishes (Pisces:Pleuronectiformes). Neth J Sea Res 5: 121–196.Ezeasor, D.N., W.M. Stokoe (1980) Scanning elec-tron microscopic study of the gut mucosa of therainbow trout Salmo gairdneri Richardson. JFish Biol 17: 529–539.Ezeasor, D.N., W.M. Strokoe (1981) Light andelectron microscopic studies on the absorptivecells of the intestine, caeca, and rectum of theadult rainbow trout, Salmo gairdneri R. J FishBiol 18: 527–544.Geevarghese, C. (1983) Morphology of the alimen-tary tract in relation to diet among gobioidfishes. J Nat Hist 17: 731–741.Harder, W. (1975) Anatomy of Fishes. Part 1(Text) and Part 2 (Figures). Stuttgart, Schwei-zerische Verlagsbuchhandlung.Hart L.J., S.D. Holladay, B.J. Smith, J. Robertson,S.A. Smith (1997) Exposure of tilapian fish tothe pesticide lindane results in hypocellularityof the primary hematopoietic organ and thespleen without altering activity of phagocyticcells in these organs. Toxicology 118: 211–221.Hossain, A.M., H.M. Dutta (1988) Embryology ofintestinal ceca in the bluegill (Lepomis macro-chirus). Can J Zool 66: 998–1003.
Gross Morphology of Tilapian Intestine Cells Tissues Organs 2000;166:294–303 303Hossain, A.M., H.M. Dutta (1992) Role of caeca asfood reservoirs in the digestive tract of bluegillfish, Lepomis macrochirus. Copeia 2: 544–547.Jacobshagen, E. (1915) Untersuchungen über dasDarmsystem der Fische und Dipnoer. Part I.Jenaische Z Naturwiss. 53: 445–556.Kapoor, B.G., H. Smith, I.A. Verighina (1975) Thealimentary canal and digestion in teleosts. AdvMarine Biol 13: 109–239.Korovina, V.M., A.V. Neyelov, Ye.P. Bondarenko(1991) The anatomy and histology of the intes-tine of the Patagonian toothfish (Dissostichuseleginoides Smitt). J Ichthyol 31: 34.Kramer, D.L., M.J. Bryant (1995) Intestine lengthin the fishes of a tropical stream: 2. Relation-ships to diet – The long and short of a convo-luted tissue. Environ Biol Fish 42: 129–141.Lagler, K.F., J.E. Bardach, R.R. Miller, D.R.M.Passino (1975) Ichthyology, ed 2. New York,Wiley.Menin, E., O.M. Mimura (1992) Comparativeanatomy of the intestine of two species of tele-ostei fish of different feeding habits. Rev Ceres39: 308–341.Mohsin, S.M. (1962) Comparative morphologyand histology of the alimentary canal in certaingroups of Indian teleosts. Acta Zool 63: 79–130.Philippart, J.-C., J.-C. Ruwet (1982) Ecology anddistribution of tilapias; in Pullin, R.S.V., R.H.Lowe-McConnell (eds): The Biology and Cul-ture of Tilapias. Manila, International Centerfor Living Aquatic Resources Management.pp 15–60.Reifel, C.W., A .A. Travill (1978) Gross morpholo-gy of the alimentary tract in ten teleostean spe-cies. Anat Anz 144: 441–449.Reifel, C .W., A.A. Travill (1979) Structure andcarbohydrate histochemistry of the intestinein ten teleostean species. J Morphol 192: 343–360.Reinthal, P.N. (1989) The gross intestine morphol-ogy of a group of rock-dwelling cichlid fishes(Pisces, Teleostei) from Lake Malawi. Neth JZool 39: 208–225.Ribble, D.O., M.H. Smith (1983) Relative intestinelength and feeding ecology of fresh water fishes.Growth 47: 292–300.Saddler, J.B., L.M. Ashley (1960) Comparative his-tology of the pyloric ceca of certain teleost fish.Walla Walla Coll Publ 27: 1–23.Sis, R.F., P.J. Ives, D.M. Jones, D.H. Lewis, W.E.Haensly (1979) The microscopic anatomy ofthe oesophagus, stomach, and intestine of thechannel catfish, Ictalurus punctatus. J Fish Biol14: 179–186.Stevens, C.E. (1988) Comparative Physiology ofthe Vertebrate Digestive System. New York,Cambridge University Press.Stroband, H.W.J., F.M.H. Debts (1978) The ultra-structure and renewal of the intestinal epithe-lium of the juvenile grasscarp. Ctenopharyngo-don idella (Val). Cell Tissue Res 187: 181–200.Suyehiro, Y. (1941) A study of the digestive systemand feeding habits of fish. Jpn J Zool 10: 1–303.Trewavas, E. (1983) Tilapine Fishes fo the GeneraSartherodon, Oreochromis, and Danakilia.London, British Museum of Natural History,p 184.Wheeler, A.C. (1985) The World Encyclopedia ofFishes. London, MacDonald, pp 320, 351.Yamaoka, K. (1985) Intestinal coiling pattern inthe epilithic algal-feeding cichlids (Pisces, Tele-ostei) of Lake Tanganyika, and its phylogeneticsignificance. Zool J Linn Soc 84: 235–261.Zihler, F. (1982) Gross morphology and configura-tion of digestive tracts of Cichlidae (Teleostei,Perciformes): Phylogenetic and functional sig-nificance. Neth J Zool 32: 544–571.