






































Iguana 13.2 b&w text


IGUANA  •  VOLUME 13, NUMBER 2  •  JUNE 2006 133CORALLUS CANINUS

With the largest known range of any currently recognized
species of boid and a fascinating and beautiful variety of

morphologies, one might logically assume that the Emerald Tree
Boa (Corallus caninus) would be among the most recognizable and
widely kept species in herpetoculture. However, it is only recently
that a broader understanding of the species has enabled prospec-
tive keepers to adapt and develop the environmental regimes nec-
essary for this highly specialized snake to thrive in captivity.

The first installment (Iguana 13(1): 37–41) proposed a set
of guidelines for the acquisition and preparation of enclosures
appropriate for housing the species. Part II presents one sound
strategy for the acquisition of healthy, viable specimens as well
as some of the more technical details associated with husbandry.

INTRODUCTION
Preparation and forethought equal to that employed during the
conceptualization, procurement, and setup of the enclosure
should now be addressed to acquiring a snake. One of the best
resources for today’s herpetoculturists is the internet. This inter-
active knowledge base provides access to reams of information,
anecdotal experiences, and live response forums linking new
keepers with those who have already had success with Emerald
Tree Boas. Judicious use of the internet can help prospective
keepers form realistic expectations, minimize the likelihood of
unpleasant surprises, and provide a generous sounding board for
those actively seeking answers to specific questions. Researchers
involved primarily with species classification and the study of
natural history have long benefited from the publication and dis-
semination of scientific information. The information-sharing
capabilities of the internet have resulted in an exponential
increase in interest and success for herpetoculturists involved
with the captive husbandry of this and other species.

SPECIMEN ACQUISITION
As with habitat selection, personal taste and practical affordabil-
ity are important factors during the selection of an animal. The
uninitiated often tend to overlook the subtleties that differenti-
ate individuals within any given species. For C. caninus, this can

H U S B A N D R Y

Emerald Gems (Corallus caninus): 
Captive Husbandry and Propagation

Part II: Acquisition, Maintenance, and Diet

Joseph M. Polanco

Photographs by the author.

Prospective purchase of an Emerald Treeboa (Corallus caninus) should
only be considered when the snake’s outward physical appearance con-
forms to criteria established for robust, healthy animals.

A healthy Emerald Treeboa (Corallus caninus) is robust, unblemished,
brightly colored, and active, coiling tightly on elevated perches and
responding quickly to stimuli.

Iguana 13.2 b&w text.QX6  6/26/06  1:20 PM  Page 133



lead to an “all emeralds are more or less equal” mindset, which
may place too much weight on the physical beauty of an indi-
vidual specimen. This is one of the most common mistakes first-
time keepers are likely to make. As with icebergs, within the
Corallus complex, what lurks beneath the surface is what needs
to be evaluated in order to make a wise decision. The tempera-
ment and physical condition of animals available on the open
market varies widely and individuals inexperienced with the
species do not have the odds in their favor. Wild-caught animals,
especially those that have not been in a stable, controlled envi-
ronment for a minimum of one year, should never be consid-
ered by any but the most seasoned of veterans.

Acclimating wild-caught treeboas is no simple undertaking.
Any reputable dealer, even those with years of experience, will
tell you that the mortality rate of imported specimens of this
species is appallingly high. Statistics indicate that more than
50% of captives of all ages die during the first year. For those
specimens that do survive the initial acclimation period, chances
of survival in the long term (> three years) are equally grim.
Although I have not been able to collect empirical data, my own
numbers reflect a survival rate of just over 65% for adult tree-
boas and just over 70% for animals acquired prior to their onto-
genetic change.

Therefore, as a rule, only captive-born and bred specimens
(subsequently referred to as CBB), purchased from well-estab-
lished, reputable sources should be considered. Where C. cani-
nus is concerned, “impulse buying” more often than not leads
to an unfortunate experience. Morphological conformation
notwithstanding, prospective acquisitions should only be con-
sidered if and when their outward physical appearance conforms
to the criteria describing a robust, healthy animal. Animals that
appear skinny, dehydrated, blemished, dull, sluggish, or gaping,
or those that are loosely coiled or not perched at all should be
dismissed out of hand. For the purposes of this article I will
restrict all future references and recommendations to those appli-
cable to CBB specimens.

The above-mentioned physical conditions are often symp-
tomatic of a variety of potentially serious pathologies. One of
the few true challenges associated with this species is that even
the most common ailments are often difficult to detect. Devoid
of compassion or sympathy for the infirm, nature provides many
species, including C. caninus, with the instinct to mask illness.
Emeralds are so adept at this practice that by the time a prob-
lem becomes obvious the individual is often beyond salvage. For
this reason, animals with a history of illness, however minor,
should be dismissed from consideration.

The previously mentioned statistics largely reflect the more
commonly available “Northern” variety of the species. Although
susceptible to the same maladies, stress-related illnesses, and par-
asitic infestations, the Amazon Basin variety appears to benefit
from two factors that may be largely responsible for the differ-
ences seen between the two populations. First, the Basin form is
widely believed by those intimately familiar with both varieties
to be hardier and more resilient. Although this disparity itself is
undocumented and remains largely unexplained, one popular
belief is that this form’s larger geographic distribution has pro-
vided for an inherently wider set of environmental tolerances.
The other more prosaic factor has been the lack of ready avail-
ability of Basins to the export trade. Local restrictions and the
logistical challenges associated with their collection have kept the
exported numbers of Basins very low for over three decades and
the prices charged for these rarely seen but highly prized exports
have tended to be very high. With the laws of supply and
demand keeping prices high, improved care at every stage of the
importation process provides better returns for importers.

Currently, a relatively strong U.S. community of well
known, reputable breeders is involved with the propagation and
sale of both Northern and Basin varieties. On the whole, these

134 IGUANA  •  VOLUME 13, NUMBER 2  •  JUNE 2006 POLANCO

Wild-caught animals, especially those that have not been in a stable,
controlled environment for a minimum period of one full year, should
never be considered by any but the most seasoned of veterans. This
individual, although superficially healthy, was dead on arrival at the
facilities of a dealer in tropical reptiles.

Blister disease can be debilitating; acquiring an animal with the intent
of nursing it back to health is a bad idea, even for an experienced
keeper.

Iguana 13.2 b&w text.QX6  6/26/06  1:20 PM  Page 134



IGUANA  •  VOLUME 13, NUMBER 2  •  JUNE 2006 135CORALLUS CANINUS

individuals happily offer support and guidance before, during,
and sometimes after the sale. This type of networking acts as an
invaluable safety net for those interested in keeping this species.
By availing themselves of the knowledge and experiences of oth-
ers, prospective keepers are able to tip the scales steeply in their
favor, drastically reducing the duration of the C. caninus learn-
ing curve.

Husbandry Basics
The essentials of effectively keeping C. caninus in captivity can
be broken down into three categories: environment, diet, and
record keeping. The first installment discussed the importance
of stabilizing environmental factors prior to the introduction of
a live animal. Although C. caninus is not an overly delicate
species, keepers can be quickly overwhelmed with health issues
when tolerances are violated.

Until such time as accurate DNA-profiles of the species
allow for a classification by which herpetologists can identify and
differentiate individual populations, accurately fine-tuning indi-
vidual environmental variables to specific animals remains
impossible. Locality data on imported specimens are non-exis-
tent, and morphological differentiation is far too imprecise to
provide even the most remote hope of accuracy. Therefore,
today’s Emerald specialists have chosen to employ a range of
temperature and humidity standards derived from averages gath-
ered from across the entire range of the species. Tables 1 & 2 list
average temperature and relative humidity levels from across the
species’ natural range for periods ranging from 4–23 years. 

I have derived the following environmental maintenance
schedule from these data. Averaging the recorded temperature
data and buffering them by 4 °F toward the median, provides a
temperature range within which any individual, regardless of
geographic origin, can thrive. Recommended daytime high and
nighttime low temperatures for captive C. caninus should be 
84 °F (DTHT) and 74 °F (NTLT), respectively.

Although some keepers use the same data to regulate rela-
tive humidity (RH) levels, years of experimentation have led me
to depart from levels found in the wild. Table 2 reflects daytime
relative humidity highs (DTRH) and nighttime relative humid-
ity lows (NTRH) averaged from eleven locations across the
species’ range. The data from which these averages were derived
were collected over periods from 2–21 years. The reason I
choose to deviate from these natural averages is the negative
effect of sustaining such high RH levels within enclosures.
Without the natural cleansing processes found in nature, black
molds tend to grow in very high concentrations. These molds
produce spores that become air-born and eventually settle on
and adhere to every porous surface in an enclosed habitat. Even
over relatively short periods, such concentrations present an
unnaturally high risk of infection or allergic reactions for both
snakes and human handlers.

Therefore, over the years, I have adopted modified RH lev-
els that have proven completely adequate, enabling me to main-

The “Northern” variety of Corallus caninus (left; this animal is from
Suriname) is more commonly available, although the Amazon Basin
variety (right) appears to be hardier and more resilient.

Table 1. Temperature (°F) and precipitation (inches) averages across the range of Corallus caninus. Portions of the species’ range are
indicated by N (= north), S (= south), E (= east), and W (= west). RF = rainfall.

Area JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

N-Hi 86 86 87 88 87 88 89 90 92 92 90 87

N-Lo 72 72 72 73 74 73 73 74 74 74 74 73

RF 8.2 5.6 6.1 8.4 11.7 11.7 9.1 7.0 3.6 3.6 4.8 7.4

S-Hi 91 91 90 90 88 88 90 92 93 93 91 90

S-Lo 72 73 72 70 66 63 61 64 69 72 72 73

RF 10.0 8.6 8.1 4.0 2.2 0.5 0.3 1.1 1.8 5.2 6.0 8.1

E-Hi 87 86 86 87 88 89 88 89 90 90 90 90

E-Lo 76 76 75 75 75 75 75 75 76 76 77 77

RF 10.5 10.0 11.8 12.0 10.5 8.2 6.4 6.5 7.5 9.1 9.8 10.2

W-Hi 88 87 87 87 86 85 85 87 88 88 88 88

W-Lo 73 73 73 73 73 72 71 72 72 72 73 73

RF 10.5 10.0 11.8 11.9 10.5 8.2 6.4 6.5 7.5 9.1 9.8 10.2

Iguana 13.2 b&w text.QX6  6/26/06  1:20 PM  Page 135



tain a healthy population of tropical species while minimizing the
excessive formation of fungal pathogens. While actual RH num-
bers for each individual facility will vary according to the funda-
mental variables of area versus air circulation, DTRH levels from
65–75% and NTRH levels from 45–55% have proven ideal.

Diet
Like that of most opportunistic ambush predators, the natural
prey of C. caninus in the wild is known to vary widely. For the
purposes of captive maintenance, we deal with three major
sources of nutrition, all rodents. Rats, mice, and Siberian Dwarf
Hamsters are most commonly used. Each species may be fed at
various stages of a snake’s development. I use the following
guidelines in my facility (species may vary, but approximate sizes
should be watched closely):

• Siberian Dwarf Hamsters (live crawlers, ± 8 g): Initiating
the natural feeding response in neonate snakes.

• Pink rats (live or frozen/thawed, ± 8 g): Neonate
maintenance once a solid feeding response has been
established.

• Fuzzy rats (frozen/thawed, ± 16 g): Neonates > 60 g.
• Rat pups (frozen/thawed, ± 25 g): Yearling snakes > 200 g.
• Weanling rats (frozen/thawed, ± 35 g): Two-year old

snakes > 350g.
• Small rats (frozen/thawed, ± 75 g): Sub-adults > 500 g.
• Medium rats (frozen/thawed, ± 150 g): Adult animals 

> 900 g.
• Large rats (frozen/thawed, ± 250 g): Adult females 

> 1500 g.

In an effort to promote optimum fitness while avoiding obesity,
which may shorten lifespan and reduce reproductive productivity,
feeding schedules and maintenance records on every individual
should be meticulously maintained. Such records are a crucial tool
in managing these snakes, given their naturally low activity levels,
which, exacerbated by captivity, are susceptible to a number of GI-
related maladies. Avoiding these illnesses is relatively easy, given
the proper tools and understanding their use.

Feeding frequency at my facility varies according to age and
sex. Younger specimens are fed most frequently, on a 10–14-day
schedule from 30 days until approximately eight months of age.
The latter frequently coincides with ontogenesis, which is signi-
fied in C. caninus by a change in ground color from neonatal
red, green, or yellow to the familiar adult shade of green.
However, this change, which is itself an energy-related process,
is not inherently tied to age.

Neonates whose GI tracts are fully engaged at an early age,
process meals and eliminate wastes on a fairly regular, predictable
schedule. Such individuals will benefit from being kept on a 10-

day feeding interval. Animals with tendencies to process meals
less quickly should be held to a longer 14-day period. Under no

136 IGUANA  •  VOLUME 13, NUMBER 2  •  JUNE 2006 POLANCO

Frozen food size references.

Live Siberian Dwarf Hamster (crawler) with dime for size reference.

Table 2. Relative humidity table (%) averaged from across the entire range of Corallus caninus. DTRH = daytime relative humid-
ity highs; NTRH = nighttime relative humidity lows.

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

DTRH 94 94 94 95 95 93 92 91 91 91 91 93

NTRH 76 78 77 78 77 72 68 66 66 67 70 76

An 18-month-old Corallus caninus on a scale for size reference.

Iguana 13.2 b&w text.QX6  6/26/06  1:20 PM  Page 136



circumstances should a neonate be fed more than two meals
without a bowel movement. A majority of neonates will elimi-
nate the waste products from its previous meal within a few
days. This waste is literally being moved out in order to make
room for the wastes generated by the digestion of newly ingested
prey. Should an animal take two meals and fail to produce a
bowel movement, one should induce elimination. This can be
most easily accomplished by encouraging simple exercise on
approximately day eight following the second meal. Allowing
the animal to crawl freely over one’s open fingers from hand to
hand for a period of ten minutes is usually enough to produce
the desired result within hours of the session.

For animals from 8 months to three years of age, a 14-day
interval between meals is appropriate. From this stage forward,
no individual should be given more than three meals between
bowel movements. Once an animal has begun eating small rats
weighing > 75 g, feeding frequency can be reduced from 14 to
18 or even 21 days. Females can remain on this schedule, but
males, having reached sexual maturity (at approximately four
years of age) should be shifted to a monthly feeding routine. The
one exception to these adult feeding schedules is made in prepa-
ration for breeding. Because cessation of feeding is common for
both males and females during different stages of the breeding
process, “padding” the energy reserves of both during the weeks
prior to the onset of the breeding season is prudent. For animals
slated for breeding in a given calendar year, I reduce the num-
ber of days between feedings to the shortest periods mentioned
previously for their respective sexes. Males will be fed (offered
food) every 21 days and females every 18 days beginning 60
days prior to introductions and lasting throughout the entire
breeding season. Many males will lose interest in feeding shortly
after an initial introduction and may not feed again for the rest
of the season. By contrast, females will usually feed ravenously
throughout the season and stop only after ovulation.

Occasionally, males that are not actively breeding will also
stop feeding during this period. For this reason, sexually mature,
non-actively breeding males should be removed from the imme-
diate area in which breeding pairs are being housed.

Acknowledgments
For general help and sharing of information over the years, and
for this particular project as well, I thank John Benz, Rolando
Burgos, Stan Chiras, Craig and Karen Clark, Frank Fusaro,
Robert Henderson, John Martin, Al Montejo, Tony Nicoli, and
the entire online Corallus community of readers and participants
at http://www.thetreeboaforum.com. Your passion inspires and
pushes the work forward.

References
Henderson, R.W. 1993. Corallus caninus. Catalogue of American Amphibians

and Reptiles (574): 1–4.

Stafford, P.J. and R.W. Henderson. 1996. Kaleidoscopic Tree Boas: The Genus
Corallus of Tropical America. Krieger Publishing Co., Malabar, Florida.

Adult animals should receive food commensurate with their size; in
this instance, an adult female boa consumes a large rat.

IGUANA  •  VOLUME 13, NUMBER 2  •  JUNE 2006 137CORALLUS CANINUS

Highly arboreal treeboas rarely descend from elevated perches, even to
consume large prey.

Iguana 13.2 b&w text.QX6  6/26/06  1:20 PM  Page 137




