The oral apparatus (OA) of the ciliated protozoan Tetrahymena thermophila consists of four ordered arrays of ciliary units. In wild-type cells, these arrays are constant in spatial organization and vary little in size except during extreme starvation. Recessive mutations at five gene loci are known to increase the size of the OA. They do this by increasing the length of the ciliary arrays, without affecting their width and often without increasing their number beyond the usual four. Comparison of the oral arrays over a large range of sizes has revealed: (1) that the lengths of the anterior two of three parallel arrays (membranelles) are rather tightly coordinated; (2) that the specific basal body configurations resulting from remodelling of the membranelles are only slightly affected by large changes in lengths of membranelles; and (3) that the third membranelle is restricted to a nearly constant length, except in the very largest OAs in which the structure is lengthened but interrupted by a gap in the middle. This gap may reveal the spatial extent of a putative zone of basal body regression. These phenomena are not specific to any of the genotypes utilized in this investigation; the effect of the mutations is to loosen quantitative restrictions and thus reveal underlying associations and constraints.

The intracellular pattern of ciliary units in the oral apparatus of Tetrahymena thermophila is useful for analysis because it is readily described quantitatively, its development can easily be observed, and it is subject to environmentally and genetically provoked variation. The organization of this system is at the same time remarkably complex and surprisingly invariant in growing wild-type cells (Williams & Bakowska, 1982; Bakowska, Frankel & Nelsen, 1982a). The nature of this organization can be probed by provoking major changes in the size of the system. Thus, severe starvation reduces the length of the ciliary arrays within the oral apparatus without greatly affecting their organization; there is evidence for alteration of organization only when the number as well as the length of these arrays is reduced (Bakowska et al. 1982a). Substantial increase in the size of the oral apparatus cannot be brought about reliably by any nutritional regimen of which we are aware, but it is elicited by any of several genic mutations. This has permitted an analysis of the response of oral patterns to an extended variation in the size of the ciliary arrays that make up the oral apparatus. In this paper, we will concentrate on the effects of increase in the size of these ciliary arrays in the absence of any increase in their number. This will also allow us to isolate analytically the unique effects of change in the number of these arrays, which is the subject of the next paper. In both papers, phenotypic changes generated by mutations are regarded as windows through which we may perceive underlying associations and constraints that guide normal development.

Stocks

All stocks used in this study were Tetrahymena thermophila of the inbred B strain (Allen & Gibson, 1973, Table 2). The wild-type cells were of stock B-2079 (20th generation of inbreeding, established in 1979). Information about the mutant stocks utilized in this and the companion investigation is summarized in Table 1. The protocol for mutagenesis was the same as described earlier (Frankel, Jenkins, Doerder & Nelsen, 1976), with the substitution of 0 · 75 % ethylmethane sulfonate (EMS) for nitrosoguanidine in one case. Mutations were brought to expression either as heterozygotes by macronuclear allelic assortment (Carlson, 1971; Frankel et al. 1976) or as homozygotes by ‘cytogamy’, i.e. induced self-fertilization (Orias, Hamilton & Flacks, 1979; Sanford & Orias, 1981). Following either of these procedures, cells were isolated into microdrops by the ‘Poisson lottery’ procedure (Orias & Bruns, 1976) and replicated by the method of Roberts & Orias (1973) into microtitre plates. Screening was for morphological abnormalities at 39 · 5°C as described by Frankel et al. (1976). Subsequent genetic analysis was carried out following procedures originally introduced by Nanney, as described in Frankel et al. (1976). Localization of mutations to chromosome arms was carried out by crosses to germinal nullisomies (Bruns & Brussard, 1981; Bruns, Brussard & Merriam, 1983), with the requisite stocks (Bruns, Brussard & Merriam, 1982) kindly provided by Dr Peter Bruns. Assessment of linkage among psm mutations was by the interrupted genomic exclusion method of Ares & Bruns (1978).

Table 1.

Mutant stocks employed in this investigation

Mutant stocks employed in this investigation
Mutant stocks employed in this investigation
Table 2.

Dimensions of cells and of OAs* in wild-type and big

Dimensions of cells and of OAs* in wild-type and big
Dimensions of cells and of OAs* in wild-type and big

Media and growth conditions

Cells were grown in one of three different peptone-based culture media described by Nelsen, Frankel & Martel (1981), with all of the analyses of mutants carried out in the 2 % proteose peptone -0 · 5 % yeast extract (PPY) medium. Temperatures of culture growth ranged from 180 to 38 °C. Procedures for growth of mass cultures were the same as described in Frankel, Mohler & Frankel (1980), except that cultures for oral isolation were shaken continuously during growth, and harvested at cell densities ranging from 4 × 104 to 2 × 105 cells per ml. Two exceptions are the 23° psmA1 culture and the 30° mpD culture (see Results), in which the flasks were not shaken and were maintained under conditions of poor temperature control, hence in these two cases the temperatures indicated are only approximate.

Oral isolations and cytology

Oral isolations were carried out following the methods described by Williams & Bakowska (1982), with use of four to six drops of SEMT (1 M-sucrose, 1 MM EDTA, 0 · 1 % 2-mercaptoethanol, 10 mM-Tris, final pH9 · 3) plus one to two drops of a 10 % Triton-X100 solution as an underlayer during centrifugal collection of oral apparatuses. All solutions were filtered carefully prior to use. Following centrifugation, the supernatant was aspirated, and the remaining pellet suspended in about two drops of distilled water, and then fixed for 2 · 5 min in 1 · 2 ml of cold 1 % OsO4. This suspension was then diluted with 50 % ethanol, centrifuged, and resuspended in 100 % ethanol. It was prepared for scanning electron microscopy as described earlier (Bakowska et al. 1982a).

Chatton-Lwoff silver impregnation of whole fixed cells and their measurements were carried out as described previously (Bakowska et al. 1982a).

Statistical analysis

Statistical procedures were carried out as described by Sokal & Rohlf (1981). ‘Model II’ regression was employed despite some reservations attaching to use of regression procedures with data such as ours, which fit the conditions described by that model (Sokal & Rohlf, 1981, section 14.13).

I. Oral organization and development in wild-type cells

The cell-surface pattern of Tetrahymena thermophila is organized around ciliary units embedded within a triton-insoluble lamina (the epiplasm) located directly underneath the cell membrane (Williams & Bakowska, 1982). Eighteen to 21 longitudinal ciliary rows are uniformly spread over the cell surface. A specialized feeding structure, the oral apparatus (OA), consisting in part of modified ciliary units spaced close together, is situated near the anterior end of the cell (Fig. 1A). The OA consists of four compound ciliary ensembles, namely one undulating membrane (UM) and three membranelles (Fig. 1B). The UM is made up of two parallel rows of basal bodies, the outer one ciliated and the inner one unciliated. The anterior two membranelles (Ml and M2 respectively) consist primarily of three rows of basal bodies. At the cell’s right end of these rows the basal bodies are displaced into a ‘sculptured’ pattern that is characteristic for each membranelle, and is virtually invariant (Bakowska et al. 1982a). The third membranelle (M3) appears to be sculptured in toto, so that the original parallel arrangement of basal body triplets is obscured. Within the membranelles, all basal bodies are ciliated except a few at the left end of Ml and the right-most basal bodies in the sculptured region of M2 and M3.

Fig. 1.

Anatomy and development of the oral apparatus of Tetrahymena thermophila. (A) A sketch of the arrangement of basal bodies on the ventral surface of a cell entering oral development. Each dot indicates a basal body; cilia are omitted. Seven ciliary rows (CR) are shown, as well as the anterior oral apparatus (OA) and a midbody oral primordium (OP). (B) A more detailed view of the arrangement of basal bodies in the oral apparatus of wild-type cells. Closed circles indicate ciliated basal bodies, while dashed circles show unciliated basal bodies. Ml, M2, and M3 are membranelles 1,2, and 3 respectively; UM is the undulating membrane. (C) Stages of midbody oral development. Each of the ten sketches shows a progressively later stage of oral development. Symbols are as in (B), with a central dot within circles indicating basal bodies of the right-postoral (stomatogenic) ciliary row, stippling of circles indicating basal bodies soon to be resorbed. The dotted lines connecting circles in the last two diagrams indicate the probable pathways of basal-body displacement during the sculpturing process. The designation of substages of stages 4 and 5 follows Lansing et al. (1984). ‘Pl’ signifies ‘pre-1’ (Nelsen et al. 1981). For further explanation, see the text.

Fig. 1.

Anatomy and development of the oral apparatus of Tetrahymena thermophila. (A) A sketch of the arrangement of basal bodies on the ventral surface of a cell entering oral development. Each dot indicates a basal body; cilia are omitted. Seven ciliary rows (CR) are shown, as well as the anterior oral apparatus (OA) and a midbody oral primordium (OP). (B) A more detailed view of the arrangement of basal bodies in the oral apparatus of wild-type cells. Closed circles indicate ciliated basal bodies, while dashed circles show unciliated basal bodies. Ml, M2, and M3 are membranelles 1,2, and 3 respectively; UM is the undulating membrane. (C) Stages of midbody oral development. Each of the ten sketches shows a progressively later stage of oral development. Symbols are as in (B), with a central dot within circles indicating basal bodies of the right-postoral (stomatogenic) ciliary row, stippling of circles indicating basal bodies soon to be resorbed. The dotted lines connecting circles in the last two diagrams indicate the probable pathways of basal-body displacement during the sculpturing process. The designation of substages of stages 4 and 5 follows Lansing et al. (1984). ‘Pl’ signifies ‘pre-1’ (Nelsen et al. 1981). For further explanation, see the text.

The OA develops from an oral primordium, which in growing cells normally appears at the cell’s left of the midregion of the right-postoral ciliary row (Fig. 1A). Stages of development of this primordium are shown diagrammatically in Fig. 1C. A field of basal bodies is produced (stage 1), from which oriented couplets of basal bodies are generated (stage 2). These couplets line up side by side to form two-rowed promembranelles (stages 3 and 4a). A third basal body is then generated anterior to each couplet (stage 4b), converting these couplets into columns of three basal bodies each. A fourth basal body is added anterior to the one or two columns at the right end of each membranelle (stage 5). These events create membranelle prototypes of nearly rectangular organization, with row- and-column organization that becomes slightly tilted to generate a hexagonal close packing. This simple form is modified near the end of stage 5 by a combination of three processes, namely ciliary regression, basal body resorption, and ciliary unit displacement. These processes generate the characteristic sculpturing of the right ends of the membranelles and a notch at the anterior-left end of Ml. This remodelling during late oral development (described in detail by Bakowska, Nelsen & Frankel, 1982b) endows each membranelle with a unique pattern signature which is distinguishable at a glance when the final structure is examined at sufficient resolution.

The UM develops quite differently from the membranelles. The details are complex (see Nelsen, 1981; Bakowska et al. 1982b ; Lansing, Frankel & Jenkins, 1984), but the essential feature of importance here is that a pro-UM develops during stage 4 by an alignment of single basal bodies at the right edge of the oral primordium, virtually orthogonal to the developing membranelles. The staggered double-row organization of the completed UM is elaborated later.

The typical sequence of midbody oral development illustrated in Fig. 1C is a prelude to cell division. The completed oral primordium becomes the OA of the posterior division product, while the OA of the anterior division product is derived from the pre-existing anterior OA. Tetrahymena also manifests an alternative mode of oral development, called oral replacement (Frankel & Williams, 1973). Here a stage-1 oral primordium is formed anteriorly, in part adjacent to the anterior end of the right postoral ciliary row and in part from the UM of the old OA (Frankel, 1969; Kaczanowski, 1976). These two basal body fields normally become fused into a single large field, from which membranelles and UM develop in the same way as in predivision oral development. However, the OA thus formed is not segregated into a posterior division product, but instead replaces the old OA, whose membranelles are resorbed. Oral replacement can function as a physiological substitute for cell division under conditions not permissive for division (Frankel, 1970; cf. Tartar, 1966; DeTerra, 1969), or as a part of sequences of morphogenetic transformation to an elongated ‘rapid swimmer’ form in Tetrahymena thermophila (Nelsen, 1978) or to ‘macrostome’ forms in certain other Tetrahymena species (Williams, 1960; Stone, 1963; Buhse, 1966; Méténier & Groliére, 1979). In the last-mentioned cases, the new oral apparatus that develops is very much larger than the old one that it replaces.

II. Mutations that increase the size of the oral apparatus

(a) Genetics

The mutations considered here, mpD, big, psmAl, psmA2, psmB, andpsmC, are nitrosoguanidine-induced single locus récessives (Table 1). All are nonallelic, except for psmAl and psmA2. psmC, psmB, and psmAl have been shown, using the methods of Ares & Bruns (1979), to be mutually unlinked. psmB was previously localized to chromosome arm 4L by Bruns (1982). Using the same methods and stocks, we have found that psmA and psmC are both on chromosome 5, while mpD is on chromosome 3R.

(b) Phenotypes

The phenotypes considered here fall into two distinct classes, that of big (and of mpD at a permissive temperature), and that of the psm family.

big is a non-conditional mutation that was selected on the basis of the unusually large size of homozygous cells (Table 2). big is different from the fat mutations isolated earlier (Frankel et al. 1976; Jenkins & Frankel, unpublished) because in big, unlike fat, cell length as well as width are substantially increased

(Table 2), as is the number of ciliary rows (not shown) and the size of the oral apparatus (Table 2, Figs 4 and 5). Cultures of big cells grow exponentially at 28 °C, with a doubling time 25 to 50% longer than that of parallel wild-type cultures; in such cultures, about 90% of the cells that are undergoing oral development are engaged in predivision development with midbody oral primordia (Figs 2, 3), while the other 10% are carrying out oral replacement. Roughly 90% of the OAs formed have 3 membranelles (Fig. 4), while the remainder have short ‘extra’ membranelles either anterior to Ml or posterior to M3. 3-membranelled OAs predominate even after a shift to 39 · 5°C, a temperature that is near the upper limit for continuous exponential culture growth.

Fig. 2.

Figs 28. Photographs of silver-impregnated T. thermophila cells, focused an oral primordia (Figs 2, 3, 6 and 7) or membranelles of mature OAs (Figs 4, 5 and 8). All of these photographs are printed at the same magnification, with the scale bar, shown in Fig. 2, indicating 10μm. A big cell with a midbody oral primordium (OP) at stage 2. The anterior oral apparatus (OA) is out of focus.

Fig. 2.

Figs 28. Photographs of silver-impregnated T. thermophila cells, focused an oral primordia (Figs 2, 3, 6 and 7) or membranelles of mature OAs (Figs 4, 5 and 8). All of these photographs are printed at the same magnification, with the scale bar, shown in Fig. 2, indicating 10μm. A big cell with a midbody oral primordium (OP) at stage 2. The anterior oral apparatus (OA) is out of focus.

Fig. 3.

A big cell with a midbody oral primordium at stage 5a. The three membranelles are not sculptured at their right ends.

Fig. 3.

A big cell with a midbody oral primordium at stage 5a. The three membranelles are not sculptured at their right ends.

Fig. 4.

The anterior portion of a big cell, with a mature OA. The three membranelles (Ml, M2, M3) and the undulating membrane (UM) are labelled. Note the sculptured right ends of the membranelles.

Fig. 4.

The anterior portion of a big cell, with a mature OA. The three membranelles (Ml, M2, M3) and the undulating membrane (UM) are labelled. Note the sculptured right ends of the membranelles.

Fig. 5.

The anterior portion of a wild-type cell, with a mature OA. Notice that while Ml and M2 are shorter than in big, M3 is approximately the same size.

Fig. 5.

The anterior portion of a wild-type cell, with a mature OA. Notice that while Ml and M2 are shorter than in big, M3 is approximately the same size.

mpD (mp = ‘membraneliar pattern’) is a conditional mutation selected on the basis of increased cell length and some change in cell shape at 39 · 5 °C. Cultures of mpD cells grow approximately as rapidly at 39 · 5 °C, and probably also at 28 °C, as do wild-type cultures. Oral development in such cultures, at either temperature, is by the standard predivision mode. At elevated temperatures, virtually all oral primordia and ensuing OAs possess four or five regular membranelles rather than the usual three (see the accompanying paper). Yet even at 28 °C, when all mpD OAs have the normal 3 membranelles, they are slightly but significantly larger than in wild-type OAs (see Table 1 of Frankel, Nelsen, Bakowska & Jenkins, 1984).

Thepsm (‘pseudomacrostome’) mutations also increase oral and cell size, but in a different manner than big and mpD. In these mutations, which are all conditional, the predominant phenotypic effect observed at the restrictive temperature is a change in longitudinal position (Frankel, 1979) and size of oral primordia. The result is generally a switch, partial or complete, from predivision to oral replacement development (Figs 6,7). The oral replacement primordia are typically unusually long (Fig. 6), and often give rise to correspondingly large OAs (Fig. 8). At the same time cell size increases, presumably because cell growth is continuing without cell division. Under appropriate conditions (e.g. psmAl at 28 °C) cultures can be maintained in which some cells are dividing while most are undergoing repeated oral replacement, so that cell number increases slowly while many individual cells become large and sometimes misshapen.

Fig. 6.

ApsmAl cell from a 28 ° culture with an anterior oral replacement primordium (OP) at stage 3. The membranelles of the old OA are to the cell’s left (viewer’s right) of the oral-replacement primordium.

Fig. 6.

ApsmAl cell from a 28 ° culture with an anterior oral replacement primordium (OP) at stage 3. The membranelles of the old OA are to the cell’s left (viewer’s right) of the oral-replacement primordium.

Fig. 7.

A psmAl cell with an anterior oral replacement primordium at stage 5a. M3 is similar in form to Ml and M2, and only slightly shorter than M2. Fragments of regressing old membranelles (RM) are visible to the cell’s anterior-left of the oral replacement primordium.

Fig. 7.

A psmAl cell with an anterior oral replacement primordium at stage 5a. M3 is similar in form to Ml and M2, and only slightly shorter than M2. Fragments of regressing old membranelles (RM) are visible to the cell’s anterior-left of the oral replacement primordium.

Fig. 8.

A psmAl cell with a mature OA. The three membranelles are labelled. Note that M3 is split into two parts.

Fig. 8.

A psmAl cell with a mature OA. The three membranelles are labelled. Note that M3 is split into two parts.

Fig. 9.

Mutual relationship of length of Ml and M2, assessed in terms of number of basal bodies, in 3-membranelled OAs from cells of (A) wild-type (□), mpD (•), and big (Δ) genotypes, all at 28°C, and (B) psmAl at 23 °C (○), psmAl at 28°C (■), psmB 2 h after a shift from 280 to 36 · 5 °C (◧), and psmC212 h after a shift from 280 to 36 · 5°C (⊙). The dotted line in both graphs is an arbitrary reference line of equality of the two variates (Ml = M2). The dashed line in (A) gives the best-fit linear regression computed for the mpD and big OAs. Note the extreme compression of the scale in (B) compared to (A). The points of (A) would fit in the shaded region of(B).

Fig. 9.

Mutual relationship of length of Ml and M2, assessed in terms of number of basal bodies, in 3-membranelled OAs from cells of (A) wild-type (□), mpD (•), and big (Δ) genotypes, all at 28°C, and (B) psmAl at 23 °C (○), psmAl at 28°C (■), psmB 2 h after a shift from 280 to 36 · 5 °C (◧), and psmC212 h after a shift from 280 to 36 · 5°C (⊙). The dotted line in both graphs is an arbitrary reference line of equality of the two variates (Ml = M2). The dashed line in (A) gives the best-fit linear regression computed for the mpD and big OAs. Note the extreme compression of the scale in (B) compared to (A). The points of (A) would fit in the shaded region of(B).

Fig. 10.

Mutual relationship of length, assessed in terms of number of basal bodies, of the UM and M2 in 3-membranelled OAs from wild-type and mutant cells. The symbols have the same meaning as in Fig. 9. The main graph (A) includes all of the data points except two, which are added to a reduced version of the same graph with extended ordinate and abscissa (B).

Fig. 10.

Mutual relationship of length, assessed in terms of number of basal bodies, of the UM and M2 in 3-membranelled OAs from wild-type and mutant cells. The symbols have the same meaning as in Fig. 9. The main graph (A) includes all of the data points except two, which are added to a reduced version of the same graph with extended ordinate and abscissa (B).

As in big, the ‘pseudomacrostome’ OAs formed by the psm mutant cells typically possess the normal complement of 3 membranelles. However, abnormalities are fairly common. These may involve an interrupted M3 (Fig. 8), short supernumerary anterior membranelle fragments, or even a tandem subdivision of all or part of the OA into two segments of similar organization but unequal size, probably due to failure of union of the two oral replacement subfields [as described by Kaczanowski (1976) in another mutant].

Of the ‘pseudomacrostome’ mutations, psmAl is the most strongly expressed, with a majority of cells in the oral replacement mode even at 28 °C; the permissive temperature for this mutation is 23 °C, or below. By contrast, psmA2 cells show no expression at 28 °C, little at 36 · 5 °C, and delayed expression following transfer from 28 ° to 39 · 5 °C. psmB andpsmC (only one allele known for each locus) are intermediate, with no expression at 28 °C and a rapid onset of pseudomacrostome-type oral replacement in 20 to 30 % of the cells following a shift to 36 · 5 °C. There is also an additional recessive, nitrosoguanidine-induced mutation, named psmD, located on chromosome 3R, that resembles the other psm mutations in sometimes bringing about formation of large OAs by oral replacement at high temperature. This mutation, unlike the other psm mutations, may also express an mp-like phenotype, with extra membranelles present even in OAs of normal size. Although expression of the two phenotypes can be dissociated, they appear to be outcomes of the same mutational lesion. OAs of this mutation have not been studied in detail because of relatively low penetrance at temperatures below 37 °C.

III. Analysis of oral patterns

(a) Coordinate regulation of the length of the first and second membranelles

Scanning electron micrographs of OAs from detergent-extracted preparations were used for assessment of the number and arrangement of ciliary units in different parts of the OA. Cilia are mostly detached during preparation at a point just distal to the basal body, leaving prominent, thick-walled stumps. Non-ciliated basal bodies have distal terminations within the surface lamina (epiplasm) and can sometimes be seen (Fig. 12). The holes in the preparation, which form rows posterior to each membranelle, are openings of perforations in the epiplasm known as parasomal sacs (Williams & Bakowska, 1982). Their visualization in these preparations is variable, even in OAs from wild-type cells. Epiplasmic ridges are located between membranelles, especially M2 and M3 (Fig. 12, r; cf. Smith, 1982). This feature has not been noted before in such preparations, but is fairly regularly seen, and is useful for distinguishing between separate membranelles and interrupted portions of a single membranelle.

The preparations of isolated OAs allow assessment of the dimensions of membranelles by counting of basal bodies. The longer dimension of the membranelles is termed ‘length’ and is tallied as number of basal body columns (indicated by numerals in Fig. 11), while the shorter dimension is the ‘width’ and is counted as the number of basal bodies per column, i.e. number of rows (indicated by the letters a, b, and c in Fig. 11). Length assessed in this manner includes the sculptured region, a convention different from that of Bakowska et al. (1982a), in which the regions of unmodified and modified (sculptured) columns were tallied separately. The basal bodies designated x and y in Fig. 11 are those of a short additional row formed anterior to row a at the right ends of the membranelles (see Fig. 1C and accompanying text). Basal body y appears anterior to column 1, while basal body x is the sole remnant of a column situated to the right of column 1 of Ml, which is resorbed before the end of oral development (Fig. 1C).*

Fig. 11.

Patterns of sculpturing of membranelles. Conventions, except for shading, are the same as in Fig. 1B,C. The numbers within the basal bodies of the top row identify the basal body columns, proceeding from the cell’s right end (viewer’s left) of each membranelle to its left. The anterior member of each set of three basal bodies (i.e. of each column) is designated a, the middle one b, and the posterior member c, creating a coordinate system for identification of basal bodies, x and y are the ‘fourth row’ basal bodies. Membranelle 3 (M-3) is shown complete, while only the right-most six columns are shown for M-l and M-2.

The diagrams on the left show the normal sculpturing patterns of the membranelles, while the central diagrams, prefixed with an ‘e’, indicate the corresponding extended patterns, with posterior displacement of certain additional basal bodies (shaded). ‘eM-2’ actually indicates three distinct patterns, since basal bodies 3b and 4c may be displaced separately or jointly (as shown). The ‘cM-2’ pattern is characterized by ciliation of the basal bodies of the first column plus basal body y (and persistence of basal body x), but otherwise resembles the M-2 pattern.

Fig. 11.

Patterns of sculpturing of membranelles. Conventions, except for shading, are the same as in Fig. 1B,C. The numbers within the basal bodies of the top row identify the basal body columns, proceeding from the cell’s right end (viewer’s left) of each membranelle to its left. The anterior member of each set of three basal bodies (i.e. of each column) is designated a, the middle one b, and the posterior member c, creating a coordinate system for identification of basal bodies, x and y are the ‘fourth row’ basal bodies. Membranelle 3 (M-3) is shown complete, while only the right-most six columns are shown for M-l and M-2.

The diagrams on the left show the normal sculpturing patterns of the membranelles, while the central diagrams, prefixed with an ‘e’, indicate the corresponding extended patterns, with posterior displacement of certain additional basal bodies (shaded). ‘eM-2’ actually indicates three distinct patterns, since basal bodies 3b and 4c may be displaced separately or jointly (as shown). The ‘cM-2’ pattern is characterized by ciliation of the basal bodies of the first column plus basal body y (and persistence of basal body x), but otherwise resembles the M-2 pattern.

Fig. 12.

Figs 12−16. Isolated OAs from cells lysed following growth in PPY. All photographs are oriented so that the cell’s left edge of the OA corresponds to the viewer’s right. The UM is thus to the viewer’s left, the membranelles to the viewer’s right, with Ml always most anterior (up) and M3 most posterior (down). Arrowheads refer to the state of ciliation of basal bodies, straight arrows indicate basal bodies displaced less than normal, while wavy arrows indicate basal bodies displaced more than normal. The membranelles are individually labelled (Ml, M2, M3) as is the anterior end of the undulating membrane (UM). The posterior portion of the UM is frequently displaced or broken off in whole or part. Scale bars indicate 1 μm. A typical OA from a wild-type cell grown at 36-5 °C. All ciliated basal bodies are visible, except for the y basal body of Ml and a few basal bodies at the left end of Ml and possibly of M2, which are covered by folds of the surface lamina (epiplasm). Most of the preparation is seen in external view, with ciliated basal bodies visible as thick-walled rings. Unciliated basal bodies of column 1 of M2 (arrowheads) are barely visible as much thinner rings. Row a and the left end of row b of Ml are seen in side view owing to the folding over of the epiplasmic border of the OA. Note epiplasmic ridges (r) between membranelles. Compare with Figs 1B and 11 (left column).

Fig. 12.

Figs 12−16. Isolated OAs from cells lysed following growth in PPY. All photographs are oriented so that the cell’s left edge of the OA corresponds to the viewer’s right. The UM is thus to the viewer’s left, the membranelles to the viewer’s right, with Ml always most anterior (up) and M3 most posterior (down). Arrowheads refer to the state of ciliation of basal bodies, straight arrows indicate basal bodies displaced less than normal, while wavy arrows indicate basal bodies displaced more than normal. The membranelles are individually labelled (Ml, M2, M3) as is the anterior end of the undulating membrane (UM). The posterior portion of the UM is frequently displaced or broken off in whole or part. Scale bars indicate 1 μm. A typical OA from a wild-type cell grown at 36-5 °C. All ciliated basal bodies are visible, except for the y basal body of Ml and a few basal bodies at the left end of Ml and possibly of M2, which are covered by folds of the surface lamina (epiplasm). Most of the preparation is seen in external view, with ciliated basal bodies visible as thick-walled rings. Unciliated basal bodies of column 1 of M2 (arrowheads) are barely visible as much thinner rings. Row a and the left end of row b of Ml are seen in side view owing to the folding over of the epiplasmic border of the OA. Note epiplasmic ridges (r) between membranelles. Compare with Figs 1B and 11 (left column).

The mpD, big, and psm mutations all increase the length of Ml and M2, while leaving their width at the standard value of 3 (Figs 1320); very rarely, a short ectopic fourth row of basal bodies is observed in OAs of psmAl (Fig. 21). The increase in length of M2 is modest and monomodal in 3-membranelled OAs of mpD cells, greater and also monomodal in OAs of big cells, and much more variable in OAs of psm cells (Table 3). InpsmAl at 28°C, the longest M2s are fourfold the usual wild-type length. Similar relationships are observed for Ml (data not shown).

Table 3.

Length of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*

Length of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*
Length of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*
Fig. 13.

An OA from a big cell grown at 28 °C. Note the similarity to wild-type in arrangement of basal bodies, despite the difference in length of Ml and M2.

Fig. 13.

An OA from a big cell grown at 28 °C. Note the similarity to wild-type in arrangement of basal bodies, despite the difference in length of Ml and M2.

When lengths of both Ml and M2 are assessed in the same OA, their mutual relationship can be evaluated. This is shown for 3-membranelled OAs of wildtype, mpD, and big cells in Fig. 9A, and for OAs of psm mutants (on a compressed scale) in Fig. 9B. Three conclusions can be drawn from these plots. First, there is a clear-cut and strong association between the lengths of Ml and M2. Second, the relationship between Ml and M2 is not influenced by the genotype of the cells: where values of M2 are the same, the values of Ml are similar irrespective of genotype (which is why all the data points could not be seen on a single plot). Third, no single linear regression fits the entire range of values of Ml and M2; even within the wild-type group, analysis of covariance reveals a significant difference of adjusted means between the distinct subset of small 3-membranelled OAs from starved cells and the remaining normal-sized OAs.

Comparison of the data points to an arbitrary reference line set at Ml = M2 (Fig. 9A and 9B, dotted line) suggests that the overall M1−M2 relationship is curvilinear. This global curvilinearity is not inconsistent with a near-linear relationship within a restricted range of values of Ml and M2; for example, a linear regression computed for the combined mpD and big data (Fig. 9, dashed line) provides an excellent fit for values of Ml between 16 and 22, but fails when extrapolated outside of this range.

(b) Imperfect coordination of the length of the undulating membrane and membranelles

The undulating membrane tends to be somewhat larger in mutant than in wildtype OAs (Table 4). However, 3-membranelled OAs of mpD and big cells have a similar range of UM lengths (Table 4) despite clearly different lengths of M2 (Table 3) and Ml (Fig. 9A). UMs of length exceeding 30 have been found only inpsmAl OAs, where they may be under-represented due to the tendency of the longest UMs to be lost in preparation.

Table 4.

Length of the undulating membrane in 3-membranelled OAs of wild-type and mutant cells*

Length of the undulating membrane in 3-membranelled OAs of wild-type and mutant cells*
Length of the undulating membrane in 3-membranelled OAs of wild-type and mutant cells*

The association between the length of the UM and that of M2 (Fig. 10) is clearly much less close than that between Ml and M2; indeed, among 3-membranelled OAs of mpD and big cells, the length of the UM is not significantly correlated with that of M2. The scanty data for the OAs of psm cells suggest a similar weak relationship between UM and M2 lengths among moderately enlarged OAs, but imply a substantial increase of UM lengths in extremely large OAs (note the two points at the upper-right corner of Fig. 10B). The difficulty of obtaining unbiased data on UM lengths makes it difficult to draw unequivocal conclusions; however, it appears as if much of the variation in UM length is generated by causes unrelated to membranelle length.

(c) Limited variation in patterns of sculpturing of membranelles

Despite substantial increases in length of membranelles brought about by the mutations under consideration, the distinctive sculptured patterns of the membranelles are affected only modestly. The normal patterns and the modifications most commonly observed in 3-membranelled OAs are shown schematically in Fig. 11 and are documented in Figs 12 to 21. In wild-type cells grown at temperatures between 18° and 37 °C (Fig. 12) and in 3-membranelled mpD cells (Frankel et al. 1984, Fig. 9) sculpturing is virtually invariant, manifesting the standard patterns shown in Fig. IB and in the left column of Fig. 11. Such normal patterns predominate in the other mutants as well (e.g. Figs 13,17), but modifications are fairly common (Table 5). The most frequent modifications, found in 3-membranelled OAs of all mutants except mpD, involve individual displacement of the 3c basal body in Ml and of the 4c basal body in M2 or M3 (Fig. 11, centre). The displacement of the 4c basal body has the effect of extending the sculptured region by one column to the cell’s left, for which reason we call it ‘extended’ sculpturing. Examples are shown in Fig. 14 for Ml, Figs 14, 19, 20, and 21 for M2, and Fig. 16 for M3. In M2, the 3b basal body may be displaced as well (Fig. 19). More rarely, a subnormal displacement of basal bodies is observed, which we call ‘reduced’ sculpturing (Fig. 15). Reduced and extended sculpturing may occur together (Fig. 16).

Table 5.

Sculpturing of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*

Sculpturing of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*
Sculpturing of membranelle 2 in 3-membranelled OAs of wild-type and mutant cells*
Fig. 14.

Another OA from a big cell grown at 28 °C in PPY. The UM is raised on a wedge of epiplasm, and its basal bodies are seen mostly in side view, obscuring most of M3. Ml and M2 show extended sculpturing, with basal body 3c of Ml and 4c of M2 (wavy arrows) displaced posteriorly; compare with the eM-1 and eM-2 diagrams in Fig. 11, in which these basal bodies are stippled.

Fig. 14.

Another OA from a big cell grown at 28 °C in PPY. The UM is raised on a wedge of epiplasm, and its basal bodies are seen mostly in side view, obscuring most of M3. Ml and M2 show extended sculpturing, with basal body 3c of Ml and 4c of M2 (wavy arrows) displaced posteriorly; compare with the eM-1 and eM-2 diagrams in Fig. 11, in which these basal bodies are stippled.

Fig. 15.

An OA from apsmB cell maintained for 2 h at 36-5 °C. The posterior half of the UM is elevated, obscuring most of M3. Ml is unsculptured, with the 1c and 2c basal bodies (arrows) not posteriorly displaced (basal bodies x and y are not visible, probably covered by an epiplasmic flap). Sculpturing of M2 is reduced, with basal bodies 2b, 2c, and 3c (arrows) displaced much less than normal (compare with Figs 12, 13).

Fig. 15.

An OA from apsmB cell maintained for 2 h at 36-5 °C. The posterior half of the UM is elevated, obscuring most of M3. Ml is unsculptured, with the 1c and 2c basal bodies (arrows) not posteriorly displaced (basal bodies x and y are not visible, probably covered by an epiplasmic flap). Sculpturing of M2 is reduced, with basal bodies 2b, 2c, and 3c (arrows) displaced much less than normal (compare with Figs 12, 13).

Fig. 16.

An OA from a psmB cell maintained for 2h at 36 · 5 °C. The sculptured pattern of M2 is highly abnormal, with simultaneously reduced displacement of basal bodies 2b, 2c, and 3c (arrows) and extended sculpturing due to anomalous displacement of basal bodies 3b and 4c (wavy arrows). In addition, basal body la is anomalously ciliated (arrowhead). The sculpturing of M3 is extended, probably due to displacement of basal body 4c (wavy arrow, compare with the stippled basal body in the eM-3 pattern in Fig. 11).

Fig. 16.

An OA from a psmB cell maintained for 2h at 36 · 5 °C. The sculptured pattern of M2 is highly abnormal, with simultaneously reduced displacement of basal bodies 2b, 2c, and 3c (arrows) and extended sculpturing due to anomalous displacement of basal bodies 3b and 4c (wavy arrows). In addition, basal body la is anomalously ciliated (arrowhead). The sculpturing of M3 is extended, probably due to displacement of basal body 4c (wavy arrow, compare with the stippled basal body in the eM-3 pattern in Fig. 11).

Fig. 17.

A psmAl OA with great elongation of Ml, M2 and UM combined with a nearly normal membranelle sculpturing pattern. The sculptured ends of Ml and M2 are almost completely visible, and normal. M3 is of normal size but somewhat abnormal pattern. The UM has been torn loose from the remainder of the preparation. The preparation is highly flattened, which may account for fading of some basal bodies located in regions where basal bodies are normally situated in depressions of the surface (open arrows). Cilia are retained at the left ends of Ml and M2.

Figs 1718. Isolated OAs from psmAl cells lysed following growth in PPY at 28 °C. The orientation of these photographs is the same as in Figs 12-16. Symbols have the same meaning as in Figs 1216; in addition, broad open arrows indicate places where basal bodies are partially obscured. Scale bars indicate 1μm.

Fig. 17.

A psmAl OA with great elongation of Ml, M2 and UM combined with a nearly normal membranelle sculpturing pattern. The sculptured ends of Ml and M2 are almost completely visible, and normal. M3 is of normal size but somewhat abnormal pattern. The UM has been torn loose from the remainder of the preparation. The preparation is highly flattened, which may account for fading of some basal bodies located in regions where basal bodies are normally situated in depressions of the surface (open arrows). Cilia are retained at the left ends of Ml and M2.

Figs 1718. Isolated OAs from psmAl cells lysed following growth in PPY at 28 °C. The orientation of these photographs is the same as in Figs 12-16. Symbols have the same meaning as in Figs 1216; in addition, broad open arrows indicate places where basal bodies are partially obscured. Scale bars indicate 1μm.

Fig. 18.

A psmAl OA with a normally sculptured Ml and an M2 with a cM-2 sculpturing pattern (cf. Fig. 11). Basal bodies y, la, lb and 1c of M2, invisible in many preparations, are all ciliated (arrowheads). This preparation is extremely flattened, with fading of basal bodies at the same positions as in Fig. 17 (single open arrows) and also at the position usually occupied by the epiplasmic flap overhanging Ml (double open arrows, also in Figs 19, 21).

Fig. 18.

A psmAl OA with a normally sculptured Ml and an M2 with a cM-2 sculpturing pattern (cf. Fig. 11). Basal bodies y, la, lb and 1c of M2, invisible in many preparations, are all ciliated (arrowheads). This preparation is extremely flattened, with fading of basal bodies at the same positions as in Fig. 17 (single open arrows) and also at the position usually occupied by the epiplasmic flap overhanging Ml (double open arrows, also in Figs 19, 21).

Fig. 19.

Figs 1921. Isolated OAs from psmAl cells lysed following growth in PPY at 28 °C. The orientation of these photographs is the same as in Figs 1218. Symbols have the same meaning as in Figs 1718; in addition, broad solid arrows indicate regions of putative ciliary-unit resorption or a membranelle-fragment left over from such resorption. Scale bars indicate 1 μm. A psmAl O A with extended sculpturing of M2 (displaced basal bodies 3b and 4c indicated by wavy arrows) and a split M3. The broad solid arrow indicates a gap between a somewhat modified version of the typical M3 pattern, with basal body la ciliated (arrowhead), and a membranelle-fragment with a regular row- and-column organization. There is no sign of fading of basal bodies on either side of the region indicated by the broad solid arrow, suggesting that it is an area in which basal bodies have been resorbed in vivo.

Fig. 19.

Figs 1921. Isolated OAs from psmAl cells lysed following growth in PPY at 28 °C. The orientation of these photographs is the same as in Figs 1218. Symbols have the same meaning as in Figs 1718; in addition, broad solid arrows indicate regions of putative ciliary-unit resorption or a membranelle-fragment left over from such resorption. Scale bars indicate 1 μm. A psmAl O A with extended sculpturing of M2 (displaced basal bodies 3b and 4c indicated by wavy arrows) and a split M3. The broad solid arrow indicates a gap between a somewhat modified version of the typical M3 pattern, with basal body la ciliated (arrowhead), and a membranelle-fragment with a regular row- and-column organization. There is no sign of fading of basal bodies on either side of the region indicated by the broad solid arrow, suggesting that it is an area in which basal bodies have been resorbed in vivo.

While expression of the extended sculpturing pattern is not mutation specific (Table 5), two other abnormalities appear to be peculiar to OAs ofpsm mutants, in particular psmAl. One of these is a variable shape of the cell’s right end of Ml (Fig. 21), while the other is ciliation of the normally unciliated basal bodies of column 1 of M2 (Fig. 11, right, and 18). Both of these abnormalities are most common in the very large OAs of psmAl cells.

Although the data are insufficient for meaningful statistical assessment, we have a strong impression that the degree and type of abnormality are not independently determined within separate membranelles of each OA; cases of extended sculpturing (Fig. 14) or reduced sculpturing (Fig. 15) in more than one membranelle of a single OA are sufficiently frequent to suggest coordination of sculpturing processes within OAs.

The frequency of abnormalities of sculpturing of M2 in 3-membranelled OAs seems to be associated with membranelle length. This is clearest in comparisons across mutants: abnormalities are least frequent in mpD, more frequent in big, and most frequent in psmAl at 28 °C (Table 5). However, assessment of correlations within mutant clones gives a more mixed picture: in psm clones there is a definite positive association of frequency of sculpturing abnormalities with size of the OA, while within the big clone there is no such significant association (data not shown).

(d) The role of ciliary-unit resorption in the patterning of the third membranelle

Membranelle 3 is characterized by a remarkably invariant arrangement of 12 ciliated basal bodies (Figs 12,13) that bears scant testimony to its developmental origin from a membranelle prototype resembling Ml and M2 (see Fig. 1C). As the length of Ml and M2 in 3-membranelled OAs increases, M3 shows no parallel increase: it either retains its standard size and pattern, as in the big OA depicted in Fig. 13, or occasionally undergoes a modest increase in size that typically is associated with extended sculpturing (Fig. 16).

Even in psmAl, very long Mis and M2s are commonly accompanied by M3s of normal size and only slightly abnormal pattern (Figs 17, 18; also fig. 4k of Frankel, 1983). However, in one half of the psmAl OAs with very long Ml and M2, M3 is also strikingly elongated, but in an unusual manner: a gap, sometimes partial (Fig. 20) but more usually complete (Figs 19, 21) is observed within M3. This gap is located adjacent to the typical M-3 configuration in all cases observed (e.g. Figs 19, 21), except the one shown in Fig. 20. The remainder of M3 on the other side of the gap sometimes shows the well-defined row- and-column organization characteristic of Ml and M2 (Fig. 19), but more commonly is variably disorganized (Figs 20, 21).

An interrupted M3 was commonly observed in silver-stained completed OAs of psmAl (Fig. 8), indicating that the pattern observed in Figs 19 and 21 is not an artifact of sample preparation for SEM. In contrast, M3 is almost invariably long and continuous in stage 4 and 5 developing oral primordia seen on the same silver preparations (Fig. 7). This indicates that the gap that appears within the mature M3 of the large psmAl OAs must be a consequence of ciliary-unit regression, rather than failure of initial development.

The increase in size of the oral apparatus (OA) of Tetrahymena thermophila brought about by any of five mutations has differential effects on spatial organization of ciliary units that are comparable to the previously reported effects of decrease of size brought about by starvation (Bakowska et al. 1982a). In both situations, the lengths of the ciliary arrays (membranelles and undulating membrane) change while their widths remain the same; the lengths of Ml and M2 are coordinately regulated while the length of the UM is less closely coordinated with that of M2 and presumably Ml; finally, the patterns of sculpturing of the right ends of the membranelles are not severely affected. The preservation of these relationships when change is in opposite directions (increase of size in one case, decrease in the other) strengthens the conclusions of the earlier study on starved cells (Bakowska et al. 1982a). These are, (1) the formation of membranelles is tightly integrated whereas development of the UM is a partially independent process, (2) modification of membranelle size takes place primarily through changes in number of basal-body couplets recruited into promembranelles, and (3) the spatial extent of sculpturing of these membranelles is largely independent of the number of couplets thus recruited. These conclusions are fully in accord with results of detailed studies on oral development in OAs of wild-type cells (Bakowska et al. 1982b) and of a ‘misaligned undulating membrane’ mutant (Lansing et al. 1984), which show that the formation and the sculpturing of membranelles take place at difference times during oral development, and that development of the undulating membrane not only differs greatly from that of the membranelles (Bakowska et al. 1982b) but also can be modified substantially while development of the membranelles remains virtually unaltered (Lansing et al. 1984).

The most novel observation in this study is of the enlarged and interrupted M3 patterns observed in several of the largest OAs from psmAl cells. This enlargement is an exception to the normal size- and pattern-constancy of M3, observed both in enlarged OAs of mutants and in diminished OAs of starved cells. The exception, however, sheds some light on how the normal constancy is achieved. In normal oral development, there is ‘evidence for occasional resorption of one or two basal body columns at the left end of M3’ (Bakowska et al. 1982b), while there is no indication of comparable resorption at the left end of Ml or M2. Thus, during the initial development of M3 an excess of basal-body columns is normally produced, which subsequently is pruned by a spatially localized resorption activity. The final pattern of M3 presumably can remain constant during starvation because what is diminished is the number of surplus ciliary units otherwise destined for elimination. Increasing the length of M3 above normal increases the surplus, which can be eliminated completely — up to a point. What is most revealing is the membranelle geometry that emerges when the capacity for elimination of surplus ciliary units is finally exceeded. The fact that M3 is then bipartite, with two surviving portions flanking a central gap, suggests that the zone of resorption activity is roughly wedge shaped, with a left as well as a right margin. This idea is illustrated schematically in Fig. 22. Its essence is that the zone of resorption activity is positionally specified in some reasonably precise manner, analogous to the localization of the ‘posterior necrotic zone’ of cell death in the chick limb bud (Saunders, 1967). This idea will be elaborated further, with additional experimental support, in the subsequent paper (Frankel et al. 1984).

Fig. 20.

A psmAl O A with normal sculpturing of Ml, extended sculpturing of M2 with displacement of basal body 4c (wavy arrow), and a partially interrupted M3. An M3-like sculpturing pattern is discernible at the right end of M3, but the region of basal body resorption (broad solid arrow) is well to its left. The portion of M3 to the left of this region is poorly organized, due only in part to tearing of the preparation in this region.

Fig. 20.

A psmAl O A with normal sculpturing of Ml, extended sculpturing of M2 with displacement of basal body 4c (wavy arrow), and a partially interrupted M3. An M3-like sculpturing pattern is discernible at the right end of M3, but the region of basal body resorption (broad solid arrow) is well to its left. The portion of M3 to the left of this region is poorly organized, due only in part to tearing of the preparation in this region.

Fig. 21.

A psmAl OA with modification of the right end of Ml, extended sculpturing of M2 with a highly displaced 4c basal body (wavy arrow), and a large gap in M3. A disorganized membranelle fragment (broad solid arrow) is seen to the cell’s left of the gap. The UM has been lost from this preparation. The ‘e’ indicates four extra basal bodies, of an ectopic fourth-row segment of M2.

Fig. 21.

A psmAl OA with modification of the right end of Ml, extended sculpturing of M2 with a highly displaced 4c basal body (wavy arrow), and a large gap in M3. A disorganized membranelle fragment (broad solid arrow) is seen to the cell’s left of the gap. The UM has been lost from this preparation. The ‘e’ indicates four extra basal bodies, of an ectopic fourth-row segment of M2.

Fig. 22.

A model for control of the size of M3. The circles indicate the basal bodies of M3 at stage 5d of oral development (see Fig. 1C), prior to sculpturing and resorption. The basal-body columns are numbered following the same convention as in Fig. 11. The stippled zone is the region within which all basal bodies are resorbed near the end of stage 5. The bracketed lengths indicate the number of basal body columns that may be formed during the initial development of M3 in OAs of (A) starved wildtype cells, (B) growing wild-type cells, (C) growing cells of mutants such as big and most psms, and (D) psmAl cells with very large OAs. In each case a ‘standard’ M3 is formed despite continuous variation in the original number of basal bodies, except when the original M3 is so long that it extends beyond the cell’s left end of the zone destined for resorption.

Fig. 22.

A model for control of the size of M3. The circles indicate the basal bodies of M3 at stage 5d of oral development (see Fig. 1C), prior to sculpturing and resorption. The basal-body columns are numbered following the same convention as in Fig. 11. The stippled zone is the region within which all basal bodies are resorbed near the end of stage 5. The bracketed lengths indicate the number of basal body columns that may be formed during the initial development of M3 in OAs of (A) starved wildtype cells, (B) growing wild-type cells, (C) growing cells of mutants such as big and most psms, and (D) psmAl cells with very large OAs. In each case a ‘standard’ M3 is formed despite continuous variation in the original number of basal bodies, except when the original M3 is so long that it extends beyond the cell’s left end of the zone destined for resorption.

This study revealed relatively little novelty in the modifications of sculpturing of the right ends of the membranelles. This was somewhat surprising, since mutants were used, and other mutations affecting membranelle patterns, mpA and mpB, bring about drastic and apparently random abnormalities of sculpturing (Frankel, 1983). Although abnormalities of sculpturing of membranelles are certainly more common among the 3-membranelled OAs of the mutants considered in this study (mpD excepted) than they are in growing wild-type cells, the difference is mostly one of frequency rather than kind. Thus, abnormalities which here are given names have previously been documented in photographs of OAs from wild-type cells: extended sculpturing through displacement of basal body 4c (and 4b) of M2 in figure 5 of Bakowska et al. (1982a), displacement of basal body 3b of M2 in figure 7, and reduced sculpturing of Ml in figures 8 and 9 of that paper. Only the cM-2 pattern of M2 (Fig. 11) and some variability in shape of Ml are new to this study, and these were both found only in extremely large OAs produced by oral replacement in psm cells.

This lack of novelty, however, obtains only when analysis is restricted to OAs that possess the usual three membranelles. When there are four or more membranelles, new sculpturing patterns appear that are not found in 3-membranelled OAs of the same genotype (Frankel et al. 1984). Since these new patterns are observed in mutant cells possessing OAs in which the length as well as the number of membranelles is increased, it is imperative to dissect the specific effects of increased number of membranelles from the background effects due to the presence of a mutation and the increased size of the individual membranelles. The present study has shown that this background effect, especially for the most informative mutant (mpD), is minimal.

The majority of the linkage analysis of psmA1, psmB, and psmC was carried out by Elaine Martel. Drawings were executed by Mary Thorson. The authors also thank Drs Anne W. K. Frankel, Stephen F. Ng, and Dennis Summerbell, as well as Mr Timothy Lansing, for their comments and criticisms. This research was supported by grant HD-08485 from the U.S. National Institutes of Health.

Allen
,
S. L.
&
Gibson
,
I.
(
1973
).
Genetics of Tetrahymena
.
In Biology of Tetrahymena
(ed.
A. M.
Elliott
), pp.
307
373
.
Stroudsburg, Pa
.:
Dowden, Hutchinson & Ross
.
Ares
,
M.
&
Bruns
,
P. J.
(
1978
).
Isolation and genetic characterization of a mutation affecting ribosomal resistance to cycloheximide in Tetrahymena
.
Genetics
90
,
463
474
.
Bakowska
,
J.
,
Frankel
,
J.
&
Nelsen
,
E. M.
(
1982a
).
Regulation of the pattern of basal bodies within the oral apparatus of Tetrahymena thermophila
.
J. Embryol. exp. Morph
.
69
,
83
105
.
Bakowska
,
J.
,
Nelsen
,
E. M.
&
Frankel
,
J.
(
1982b
).
Development of the ciliary pattern of the oral apparatus of Tetrahymena thermophila
.
J. Protozool
.
29
,
366
382
.
Bruns
,
P. J.
(
1982
).
Tetrahymena thermophila
.
Genetic Maps
2
,
178
181
.
Bruns
,
P. J.
&
Brussard
,
T. B.
(
1981
).
Nullisomie Tetrahymena: eliminating germinal chromosomes
.
Science
213
,
549
551
.
Bruns
,
P. J.
,
Brussard
,
T. B.
&
Merriam
,
E. V.
(
1982
).
In vivo genetic engineering in Tetrahymena
.
Acta Protozool
.
22
(
Supplement
),
31
44
.
Bruns
,
P. J.
,
Brussard
,
T. B.
&
Merriam
,
E. V.
(
1983
).
Nullisomie Tetrahymena. II. A set of nullisomies define the germinal chromosomes
.
Genetics
104
,
257
270
.
Buhse
,
H. E.
Jr
. (
1966
).
Oral morphogenesis during transformation from microstome to macrostome and macrostome to microstome in Tetrahymena vorax strain V2 type S
.
Trans. Amer. Microsc. Soc
.
85
,
305
313
.
Carlson
,
P. S.
(
1971
).
Mutant selection in Tetrahymena pyriformis
.
Genetics
69
,
261
265
.
De Terra
,
N.
(
1969
).
Differential growth in the cortical fibrillar system as the trigger for oral differentiation and cell division in Stentor
.
Expl Cell Res
.
56
,
142
153
.
Frankel
,
J.
(
1969
).
Participation of the undulating membrane in the formation of the oral replacement primordium of Tetrahymena pyriformis
.
J. Protozool
.
16
,
26
35
.
Frankel
,
J.
(
1970
).
The synchronization of oral development without cell division in Tetrahymena pyriformis GL-C
.
J. exp. Zool
.
173
,
79
99
.
Frankel
,
J.
(
1979
).
An analysis of cell-surface patterning in Tetrahymena
.
In Determinants of Spatial Organization
(ed.
S.
Subtelny
&
I. R.
Konigsberg
), pp.
215
246
.
New York
:
Academic Press
.
Frankel
,
J.
(
1983
).
What are the developmental underpinnings of evolutionary changes in protozoan morphology?
In Development and Evolution
(ed.
B. C.
Goodwin
,
N.
Holder
&
C. C.
Wylie
), pp.
279
314
.
Cambridge
:
Cambridge University Press
.
Frankel
,
J.
,
Jenkins
,
L. M.
,
Doerder
,
F. P.
&
Nelsen
,
E. M.
(
1976
).
Mutations affecting cell division in Tetrahymena pyriformis. I. Selection and genetic analysis
.
Genetics
83
,
489
506
.
Frankel
,
J.
,
Mohler
,
J.
&
Frankel
,
A. W. K.
(
1980
).
The relationship between the excessdelay phenomenon and temperature-sensitive periods in Tetrahymena thermophila
.
J. Cell Sci
.
43
,
75
91
.
Frankel
,
J.
,
Nelsen
,
E. M.
,
Bakowska
,
J.
&
Jenkins
,
L. M.
(
1984
).
Mutational analysis of patterning of oral structures in Tetrahymena. II. A graded basis for the individuality of intracellular structural arrays
.
J. Embryol. exp. Morph
.
82
,
67
95
.
Frankel
,
J.
&
Williams
,
N. E.
(
1973
).
Cortical development in Tetrahymena
.
In Biology of Tetrahymena
(ed.
A. M.
Elliott
), pp.
375
409
.
Stroudsburg, Pa
.:
Dowden, Hutchinson & Ross
.
Kaczanowski
,
A.
(
1976
).
An analysis of mp gene affected morphogenesis in Tetrahymena pyriformis, syngen 1
.
J. exp. Zool
.
196
,
215
230
.
Lansing
,
T. J.
,
Frankel
,
J.
&
Jenkins
,
L. M.
(
1984
).
Oral ultrastructure and oral development of the misaligned undulating membrane mutant of Tetrahymena thermophila
.
J. Protozool., submitted
.
Méténier
,
G.
&
Groliére
,
C. A.
(
1979
).
Précisions temporelles sur les processus de la stomatogenése de Tetrahymena paravorax analysés en microscopie photonique
.
J. Protozool
.
26
,
75
82
.
Nelsen
,
E. M.
(
1978
).
Transformation in Tetrahymena thermophila: development of an inducible phenotype
.
Devl Biol
.
66
,
17
31
.
Nelsen
,
E. M.
(
1981
).
The undulating membrane of Tetrahymena: formation and reconstruction
.
Trans. Amer. Microsc. Soc
.
100
,
285
295
.
Nelsen
,
E. M.
,
Frankel
,
J.
&
Martel
,
E.
(
1981
).
Development of the ciliature of Tetrahymena thermophila. I. Temporal coordination with oral development
.
Devi Biol
.
88
,
27
38
.
Orias
,
E.
&
Bruns
,
P. J.
(
1976
).
Induction and isolation of mutants in Tetrahymena
.
In Methods in Cell Physiology
(ed.
D. M.
Prescott
), Vol.
13
, pp.
247
282
.
New York
:
Academie Press
.
Orias
,
E.
,
Hamilton
,
E. P.
&
Flacks
,
M.
(
1979
).
Osmotic shock prevents nuclear exchange and produces whole-genome homozygotes in conjugating Tetrahymena
.
Science
203
,
660
663
.
Roberts
,
C. T.
&
Orias
,
E.
(
1973
).
A cycloheximide-resistant mutant of Tetrahymena pyriformis
.
Expl Cell Res
.
81
,
312
316
.
Sanford
,
Y. M.
&
Orias
,
E.
(
1981
).
Phenylketonuric Tetrahymena: phenylalanine hydroxylase mutants and other tyrosine auxotrophs
.
Proc. natn. Acad. Sci., U.S.A
.
78
,
7614
7618
.
Saunders
,
J. W.
(
1967
).
Death in embryonic systems
.
Science
154
,
604
612
.
Smith
,
H. E.
(
1982
).
Oral apparatus structure in the microstomal form of Tetrahymena vorax
.
Trans. Am. Microsc. Soc
.
101
,
36
58
.
Sokal
,
R. R.
&
Rohlf
,
F. J.
(
1981
).
Biometry
(2nd Edition).
San Francisco
:
Freeman
.
Stone
,
G. E.
(
1963
).
Polymorphic properties of Tetrahymenapatula during growth in axenic culture
.
J. Protozool
.
10
,
74
80
.
Tartar
,
V.
(
1966
).
Fission after division primordium removal in the ciliate Stentor coeruleus and comparable experiments on reorganizers
.
Expl Cell Res
.
42
,
357
370
.
Williams
,
N. E.
(
1960
).
The polymorphic life-history of Tetrahymena patula
.
J. Protozool
.
7
,
10
17
.
Williams
,
N. E.
&
Bakowska
,
J.
(
1982
).
Scanning electron microscopy of the cytoskeletal elements in the oral apparatus of Tetrahymena
.
J. Protozool
.
29
,
382
389
.
*

It is possible that this transient basal body column of Ml corresponds to the column numbered 1 in M2 and M3, in which case the column-designations for Ml should be increased by one. This is not done here because of indications of short-lived basal body columns at the right end of developing M2 and M3 as well (Bakowska et al. 1982b; Lansing et al. 1984).

In the preceding study (Bakowska et al. 1982a), a single regression line was drawn through both sub-groups (Fig. 6A). Reanalysis of the expanded data-set that is now available indicates that a single straight line may not be justified.