The mimic is afforded the greatest amount of protection by possible 
predators if its pattern closely resembles the model.  Therefore, selection 
favors the mimetic pattern that most closely resembles the model species.  However, the 
more the mimic and unpalatable model resemble each other the greater the chance of 
possible predation upon the model.  This is a result from predators learning that the 
aposematic signal is not always associated with lethal consequences.  Concurrently,
 selection towards differentiation from the pattern of the most abundant mimic is 
occurring in the model species – an antagonistic effect of Batesian mimicry.

     Sympatry between model and mimic is typically considered a critical 
condition of the Batesian mimicry system that, if absent, should decrease the 
protection afforded the mimic.  Many kingsnake species, however, including
L. zonata, are distributed outside the range of coral snakes.

Avian migration could be the key mechanism that perpetuates Batesian mimicry
where the model and mimic are allopatric.  Several predatory bird species such
as kites, eagles and hawks, annually engage in latitudinal migration routes
from Mexico north through California's coastline, a route beginning within coral
snake range and ending within the range of L. zonata.  

Below is the geographic distribution of L. zonata and all of its previously 
recognized subspecies in different colors.  A phylogeographic study using mtDNA
showed that the subspecific taxa of L. zonata are not genetically distinct and invalid 
(Rodriquez-Robles, 1999).  As a result of the study the seven recognized subspecies of
 L. zonata were dissolved, resulting in a single species with no subspecies.  However, I have provided 
the historical range map of the seven subspecies which depicts the overall geographic range 
of L. zonata in its current monophyletic distribution.


  

Whether avoidance of coral snake patterns stems from innate recognition of 
aposematic signals or is a learned behavior, avian predators should avoid the 
signal as a direct result of spending approximately six months each year within
the range of a lethal model.  Failure to distinguish between a model and mimic
could be lethal.  However, not all populations of L. zonata display a highly
mimetic pattern.  A trade-off may exist between mimetic resemblance and 
physiological functions; populations that inhabit high elevations have patterns
severely reduced in red, and instead have expanded black bands that may
increase thermoregulatory efficiency.  Below are three examples of pattern that 
demonstrate the variation between populations of L. zonata.
  


Highly confluent (L. zonata – average coast variant)
  


Poor confluency (L. zonata – average Sierra variant)
  


Lacking all red pigmentation (L. zonata – atypical Sierra variant)
  

From field collected and in-lab breeding data gathered over the last five years there are 
significant differences in the average amount of triad confluency (i.e., where a black ring 
is completely interrupted by a red band dorsally) and the amount of black pigmentation
between regionalized populations of L. zonata.  The Coast populations have the 
least amount of black pigmentation and are highly confluent, while the Sierra populations, 
high elevation populations, have the greatest amount of black pigmentation 
(63.67% average) of all L. zonata I have examined.  This high elevation pattern may be necessary for 
increasing thermoregulatory efficiency because of shorter seasons and overall cooler temperatures.  
However, as the amount of black pigmentation increases, the strength of the mimetic 
signal is likely compromised.  Thus, the anti-predatory advantage arising from 
mimicry, and in thermoregulatory efficiency arising from a darker-colored
dorsal pattern, may create an evolutionary trade-off.  Hypothetically then, coast populations are 
mimetic, whereas high altitude populations of L. zonata, thermoregulation has maybe
become the dominant pressure, outweighing mimicry.

Different species and subspecies of tricolor snakes (the majority being from the genus Lampropeltis)
range throughout many of the lower forty eight states.  Many do not share the same 
geographical distribution or overlap in range with any of the three coral snake species 
native to the United States.  Therefore, these snakes are phenotypically maladaptive – they
are highly conspicuous, brandishing warning coloration and having no type of secondary
 defense to back it up (i.e. venom).  The goal of my research is to explain the distribution of 
North American tricolored colubrids with ranges outside the range of coral snakes.