Transition Studies in Flowering Plants  

by Rolf Baumberger

Excerpt  (unfinished  manuscript env.  2003  no Summary, no Discussion section) 

The mischievous play of Monkeyflowers  a prelude to speciation?  

 by Rolf Baumberger & Thomas W. Baumann


The German «Gauklerblume» for the many monkey flowers assembled in the genus Mimulus signifies a bloom that has, semantically speaking, the ability to create and rapidly change illusions by behavior playfully. The present study deals with the sticky monkey flower Mimulus aurantiacus Curtis and its allies of uncertain taxonomic range and thus named in the following 'entities' (for a review see (Thompson, 1993; Tulig, 2000)). Their trickster nature has two facets. On the one side, they are known to produce at the same plant flowers that look not the same; e.g. «Australis» and «Longiflorus» entities happen to bring forth at the early blooming reddish-colored, small-sized flowers, whereas the late blossom may consist of orange or yellow flowers distinctly larger than the former. On the other side, generations of field botanists browsing during the last 165 years from South Western Oregon through the Coastal Sage Scrub, the Chaparral, and the Woodland until the Southern Oak Woodland communities of California, not only found uniform and isolated populations of Sticky Monkeyflower but also frequently noticed a geographical overlapping of two (or seldom three) entities with a complex mixture of appearances along the borders or zone of «collision». Beeks (Beeks, 1962) called this spatial phenomenon as «intergradation zone» and, not fully aware of Sticky Monkeyflower phony behavior. He regarded this zone being the temporal result of crosses and backcrosses. Therefore, he also designated it «hybridization zone.» However, the many inconsistencies of and contradictions within the concept of «hybridization zone»  led us to a completely different view not considered by anybody before. Originally, the idea was triggered (sparked) off by the pronounced «flower-visiting pressure» exerted by hummingbirds spreading out from Mexico along the coast towards the north. In short, the present study reveals that in such zones where either «Longiflorus» or «Australis» entities meet up with «Puniceus», individual plants do gradually shift within a few years from one state («Longiflorus» or «Australis» ) to the other («Puniceus»), i.e. from a yellow–flowering, insect-pollinated into a red–flowering, bird-pollinated plant, always in this sense as observed so far. Needless to say that a change of floral morphology is an important element of this «transition process,» as we call it. 


Moreover, the transition zone moves in distinct directions, leaving behind it stable monkey flower individuals showing the hummingbird pollination syndrome.


From «hybrid zone» to transition zone
During a doctoral thesis Beeks (Beeks, 1962) thoroughly explored the distribution of Mimulus aurantiacus in the coastal region of Southern California, and by doing so, he came across a zone he called «intergradation zone.» In 1962, he published his extensive data collected as early as in 1955 and reported on the geographic ranges of the three Mimulus taxa Puniceus, Australis, Longiflorus, at that time regarded as species and affiliated to Diplacus (for taxonomic definition see the section  «Material & Methods»). However, the central element was the accurate description of the route of his «intergradation zone,» as depicted in Fig 1A just below. 

It ran through the very (deep) south of California and paralleled in its lower part the coast at a distance of 25 km. To the north, in Riverside and Orange County, it has two side branches, one going coastward and the other inland. Beeks was convinced – and many Mimulus specialists share his conception (opinion) – that the populations within this entire zone emerged, as already mentioned above, from hybridization (and subsequent introgression) between two subspecies, i.e. in the south between the red-flowered Puniceus (inset A; Fig 1) of the coast and the yellow-flowered Australis (inset B; Fig 1) of the inland, whereas further north Puniceus met with the yellow-flowered Longiflorus (inset C; Fig. 1). Phenomenally, except for the fork in the north, Beeks recognized along the full length of his zone flowers that exhibited all shades and shapes of in-between the neighboring «parental extremes» (pure phenotypes).


This was the state of knowledge when we visited Beeks' hotbed of «hybridization» for the first time. In 1995 and 1996, we combed his entire zone, ca 170 km in length and 2 to 10 km in width, from the very south of San Diego County until the north end. After inspection of many populations, we became doubts about Beeks' accuracy, even though he had indicated the exact coordinates for some populations on well-defined transects. However, the more populations we visited, the more apparent it became that an inexplicable change had occurred, this especially pronounced in the San Diego County where the «intergrading» populations showing the «hybrid» floral pigmentation described by Beeks were regularly found 3 to 10 km farther inland of his zone. Not enough, in the center of «Magic Park» (Santa Rosa Plateau) the yellow populations depicted by Beeks between Lake Elsinore and Temecula were «absent» already in 1996 and «replaced» by uniformly red ones.

Fig1B just below.

click on Figure 1B to see the legend

Plants in the «intergrading zone» bloom three weeks earlier than the pure yellow form not far from them and flowers develops in a solid burst to produce an orange spectacle obviously well remarked by hummingbirds, which migrate in April through these regions (Grant & Grant, 1968). Moreover, the size of populations had increased during our monitoring by a factor of 2 to 3. All these observations sparked off suddenly the idea, which the «intergrading zone» is rather a «transition zone» which migrates. Since we have started labeling individual plants as early as in 1997 (see below), clear evidence was collected till 2000, that not the transient population per se but the appearance of it migrated like a wave and at a speed of almost one and a half km per annum in the average, with the result that in 2003 nearly 70 % of all transient populations were located outside Beeks' zone, that is to the east of it as shown in Fig. 1B. 

Farther south near the Mexican border, we observed at the beginning of our studies in 1995 a very narrow transition zone. Between 1998 and 2003 it faded away with the result of distinct populations of red Puniceus and yellow Australis in close (ca 3 km) vicinity without intermediate forms between them.

Fig2 just below

click on the map to see the legend

See Transition SPEED 1996-2010 -->

Figure 2 summarizes the dynamics of transitions between 1997 and 2003, initially evolved from the coherent, long-stretched zone of Beeks. Two transition zones were formed, one in the North with transient Longiflorus(?)(within Camp Pendleton till San Juan Capistrano) or transient Australis (from Laguna Beach along San Juan Canyon till ca 6 km east of San Juan Hot Springs) and one in the east with transient Australis exclusively. However, the dynamics not only affect the particular shape of «geographic structure» but also include a process of «retraction» The latter could be shown unambiguously for the southern part of the zone in the east and the northern extension of the zone in the north. Changes in «geographic structure» and the retraction will go on and presumably lead to a complete dissolution of the transition zones in both regions. It has to be mentioned that artificial Longiflorus populations laid out in San Diego County were not recorded.


Transition and conversion in detail

Within the geographic limits of natural distribution of the three entities (Puniceus, Australis, and Longiflorus) several situations can be met (Fig. 2).

 • Firstly, there are red Puniceus populations near the coast reaching from the town Oceanside down to San Diego (A), yellow Australis populations in the very east (B) as well as yellow Longiflorus  populations in Orange County (C), which all kept stable as long as for 8 yrs with the one Puniceus exception in Encinitas. In Figure 2 they appear as small red and yellow circles respectively, of the same color and diameter in the years 1996 and 2003 indicating that both the color and the (slight) color standard deviation of the flowers did not change within the period of observation.

• Secondly, the population of type T1 signifies the «successful» destabilization of the yellow phenotype (Australis),  what may lead within five years to an appearance described by Beeks as «intergrading» which is concerning coloration highly significant***. However, some of the T1 populations show within five yrs only the very beginning of destabilization. Interestingly, these individuals in transition do not or only slightly flit

• Thirdly, populations can be seen (T2) with an enhanced color standard deviation indicative of a specific but limited tendency of its members to produce flowers with orange-red hues, this combined with the process of flitting     (Gaukeln): newly formed flowers exhibit an orange or already reddish pigmentation that gradually fades during their lifetime. 

Allegorically and referring to the title of this publication, the individual plant starts playing the same new motif on blooming flowers, but each time returns to the old melody. 

This stage is leading over to T3 (see next).

• Fourthly and finally, in the T3 populations, the flitting behavior as described for T2  is likewise present, however, more distinct in the sense that not only the flitting range gets narrower but also that the starting color was gradually shifting to red from one year to the other. Eventually, these T3 plants switched almost entirely to the new motif within the time of observation and remained red.

In essence, the processes of transition and final conversion are as follows: yellow-flowering populations (B or C) in vicinity of red-flowering populations (A) are «destabilized» (T1) and become significantly transient after 4-5 years (T2), altogether forming a transition zone, the appearance of which moves as described above in Fig. 1. The ultimate «goal» of this process obviously is the formation of red-flowered populations (T3) with a morphology considerably different from the yellow «ancestors'» (see QuickTime movie in low quality only [1]) and displaying the syndrome of bird pollination.

Single-plant monitoring

It is conceivable that the conventional view challenges our unexpected findings. To unambiguously prove transition and complete conversion, individual yellow-flowering plants were labeled in the field and periodically monitored over the last seven years (squares in Fig 2). The idea was that some of these marked plants would become orange-flowering within the seven years of monitoring. Indeed, the transition happened step by step, the first individual turning into the orange stage was observed in 1997 at Mount Palomar, followed by four plants near West Frontage Rd in 1998, by another four at Rainbow Canyon Rd in 1999 and by one at Santa Ysabel in 2002. Three of the four populations are located within the transition zone 2003. The fourth population of yellow  Longiflorus near West Frontage Rd had been artificially planted in connection with a restoration program in the nineties and resides now in the 'red zone.' Ten plants marked in four different populations have undergone a flower color shift from yellow to orange. There is no doubt that the individual plant turns both color and shape of its flowers from insect into hummingbird features. Interestingly, also populations planted into vicinity (ca 350 m) of red populations shifted orange. The degree of orange color achieved by all labeled plants (and thus by the population) within the time of monitoring is highly significant and until now (so far) there are no signs of a relapse into the old condition. The populations are between 10 and 60 km apart from each other.



Material & Methods


Taxonomy and the naming of Mimulus (formerly Scrophulariaceae) has recently been altered. In using molecular markers (Beardsley & Olmstead, 2002) gained evidence for placing Mimulus into the Phrymaceae.

Today, the taxon Mimulus aurantiacus signifies the entire complex consisting of at least 9 subspecies (Thompson, 1993). In this study we decided to adopt the nomenclature of Thompson published in “The Jepson Manual”, considering the three entities in question as subspecies of M. aurantiacus, which correspond to Beeks' taxa as follows: Mimulus aurantiacus ssp. puniceus = Diplacus puniceusMimulus aurantiacus ssp. australis = Diplacus longiflorus, San Diego race; Mimulus aurantiacus ssp. longiflorus = Diplacus longiflorus, northern race. Moreover, to facilitate reading we use 'Puniceus''Australis', and 'Longiflorus' when referring to subspecies or populations. Since the terms 'hybrid zone' as well 'hybrids' are limited to sexual interactions between Mimulus entities and does not comply with the gradual character shift of individual plants demonstrated here to be induced by (possibly) external factors, we use the terms 'transition zone' and 'plants in transition' instead.

Field Surveys


During 14 field trips (29 wks) in April and/or May from 1995 till 2003 as well as in January from 1999 till 2001 a total of 221 populations of Mimulus aurantiacus – PuniceusAustralisLongiflorus, and the related intermediates – were recorded in the Southern Californian Counties of Los Angeles, Orange, Riverside, and San Diego. A selection of 47 populations was scored three times, namely first in 1996 (April/May), in 1999 (April) and then again in 2003 (April/May). The sites were found along roads as well as foot trails and mapped using a mapping-grade GPS unit. At each site twenty recently opened flowers from twenty different individuals were evaluated. Practically, flower collection was made from 30 plants, one after the other, along a virtual straight line. Thereafter, 20 flowers were randomly removed from the collection bag, and of each the color of the lower corolla lobe was assessed by comparing it with the most saturated tints of the color chart of the Methuen Book of Colour (Kornerup & Wanscher, 1983) and thus yielded a color value in Methuen units. The related coordinates (L*a*b) are shown in Table 1. As flower color is the most conspicuous distinctive feature (Beeks, 1962), it was used as a transition zone marker. The criterion for transition was a mean value between 5.05 and 9.1 Methuen units of 20 flowers (tenets) collected from one and the same population. Significance of median flower color changes by time at a given site was determined by the kruskal wallis?? test, and the related pvalues were assessed by chi-square distribution. Furthermore, from our long-term observations we know that flower color, if changing to red, increases by 0.4 to 0.5 units per year at the most dynamic spots and thus, a real shift to higher Methuen units would be recognised in a population after two years at the earliest.


Single Plant monitoring


In order to follow over years any changes of the color and morphology of both the flowers and vegetative parts, a total of 60 yellow-flowering plants out of 4 populations were labelled by a (white) nylon badge (2 mm x Ø 25 mm) with a central hole to carry an insulated coper-wire ring loosely attached to the plant stem. The code was burned into the nylon with a soldering iron.


Pigment analysis


Fresh tissue samples of 100 mg (fresh wt) were homogenized in 0.5 ml chilled MeOH/0.1% HCl using a 1-ml-Eppendorf tube and a conically shaped chromesteel pistil. After centrifugation for 5 min at 4° C and 13,000 rpm the supernatant was transferred into a HPLC vial. Chromatography was carried out using a Hewlett-Packard HPLC (HP 1090) equipped with a diode array detector set at 280, 325 and 510 nm and a RP column (C18; Nucleosil, 250 x 4.6 mm). Separation of pigments was achieved following the method (modified) of (Wilbert et al., 1997) using H 2O [A] and MeOH [B], both with 0.1 % TFA, at a total flow rate of 1 ml min-1 by the following gradient (% B over A): 0-5 min (30); 5-30 min (30-46); 30-35 min (46). Injection volume was 2-50 µl. Peak identification (anthocyanidin and possible acylation) by UV/visible absorption spectra (200-600 nm) was based on (a) literature and chromatography of anthocyanins of Mimulus cardinalis, (b)  comparison with chromatograms of the 'fruit' peel of Fragaria vesca and the petals of Papaver rhoeas (pelargonidin 3-glycosides), the petals of Salvia coccinea (acylated pelargonidins) and (c) the data published by (Rodriguez-Saona et al., 1998).

For in situ pigment localisation, handcuts of petals were immersed in water and inspected by light microscopy.


Beardsley PM &  Olmstead RG, 2002: Redefining Phrymaceae: the placement of Mimulus, tribe Mimuleae, and Phryma; American Journal of Botany 89, 1093-1102.

Beeks RM, 1962: Variation and hybridization in Southern California popoulations of Diplacus (Scrophulariacea); Aliso 5(2), 83-122.

Grant KA &  Grant V, 1968: Hummingbirds and their flowers. New York, London, Columbia University Press.

Kornerup A &  Wanscher JH, 1983: Methuen handbook of colour. London, Methuen.

Rodriguez-Saona L, Giusti M &  Wrolstad R, 1998: Anthocyanin pigment composition of red-fleshed potatoes; Journal of food science 63(3).

Thompson DM, 1993: Mimulus. in The Jepson Manual: Higher Plants of California, pp. 1037-1046,  JC Hickman, ed.; University of California Press, Los Angles.

Tulig M,2000: Morphological variation In Mimulus section Diplacus (Scrophulariaceae). Department of Biological Sciences. Pomona, CA, California State Polytechnic University, 82.

Wilbert SM, Schemske DW &  Bradshaw jr HD, 1997: Floral anthocyanins from two monkeyflower species with different pollinators; Biochemical Systematics and Ecology 25(5), 437-443.

[1] The phenotypic changes during transition and complete conversion include the following elements, in this order ('pigmentation' stands for reddening by anthocyanins as described below): pigmentation of pedicel and later calyx; whole flower gets smaller; lower portion of corolla tube pigmented; increase of tube angle to the rest of the flower; pigmentation of corolla lobes; flower size diminishes again and stigma become exerted; pedicel elongates. Plant itself becomes sturdy and has smaller leaves. Anthocyanins: At the very beginning, when the first signs of reddening were visible on the petals of Australis, only cyanidins could be detected, later pelargonidins were predominant (ca 75 %) and the rest were cyanidins as based on the absorption maxima at 504 and 525 nm respectively.


Shifting borders between subspecies of bush monkeyflower Mimulus aurantiacus in Southern California.
Draft Mimulus 2002, unpublished Rolf Baumberger Zürich & Kevin Cummins San Diego
DraftMimulus2002.edited.pdf (114.58KB)
Shifting borders between subspecies of bush monkeyflower Mimulus aurantiacus in Southern California.
Draft Mimulus 2002, unpublished Rolf Baumberger Zürich & Kevin Cummins San Diego
DraftMimulus2002.edited.pdf (114.58KB)