Orchids Under Glass - Daniel Geiger's Homepage

Orchids Under Glass 

A Custom Terrarium with Bells and Whistles 

text and photographs by daniel l. geiger

I PREVIOUSLY REPORTED ON BASIC 

considerations regarding growing orchids 

in a terrarium (Geiger 2008). After a 

couple years, I noticed some aspects that 

were not ideal: 

♦ Overall size too small (that’s always 

true, isn’t it?) 

♦ Lights on top of terrarium inconvenient 

♦ Cooling during hot summer days 

with frozen water bottles problematic. 

♦ Fogging and water droplets on 

front glass 

♦ Humidity control not sufficiently 

adaptable 

♦ Too much clutter with cables and 

tubing. 

the neW terrariUM As I 

have a particular spot in our house for 

a terrarium, and because I wanted some 

special design options, I had a custom 

enclosure built through a local aquarium 

store. The first question was regarding 

the material of the tank. Glass is overall 

cheaper than acrylic and does not scratch 

as easily, but it can break, and it is more 

difficult to drill holes for tubing. Acrylic 

is more expensive, scratches more easily, 

but does not break, and one can drill 

holes easily. For the front, glass would be 

more suitable, but for the sides, where I 

wanted some tubing and equipment ports 

installed, acrylic would be more advantageous. 

Unfortunately, mixing acrylic 

and glass is not advisable, because the 

junctions of those two materials are not 

stable. I went with an all-glass tank 63 

inches long × 24 inches high × 14 inches 

deep (157 × 60 × 35 cm). The planning of 

the internal layout and all the associated 

tubing and shop-drilled holes were done 

with scale plans drawn on the computer 

in the graphics application, InDesign. 

Special design elements include 

drilled holes for line-in and line-out for 

a mister, automatic top-off system for 

waterfall/swamp cooler and cooling of 

swamp cooler. An internal compartment 

for the reservoir and a rail system to 

accommodate a piece of rock for the 

waterfall swamp cooler was installed at 

an angle. This geometry hides the equipment 

of the system in the rear portion of 

the triangular compartment. 

[1] the terrarium in the author’s dining 

room with the t5ho light suspended 

above. behind the black cloth is the 

herpKeeper control system (see page 

36 for details.) in the background is a 

4- × 9-foot (1.2- × 2.7-m) redwood-andglass 

santa barbara greenhouse.

2 

3 

4 

Integrating the water reservoir for the 

mister and top-off system into the main tank 

is not advisable, because the entire enclosure 

would have to be built to withstand the water 

weight in a relatively small portion of the 

overall tank. The glass has to be significantly 

thicker, meaning a heavier and more 

expensive tank. 

With 20–20 hindsight, drilling holes 

through glass is not as difficult as it may 

sound. For the smaller reservoir tank I dared 

drilling my own holes. A decent hardware 

store usually has glass drill bits. With an 

electric drill, a steady hand, a bit of patience 

and a water spray bottle, a hole can be drilled 

in about five minutes. Given that a builder 

charges $20 per hole, the drill bit pays for 

itself with the second hole. Tempered glass 

cannot be drilled, as it will shatter. 

Misting After trying several misters, 

I found the MistKing (www.mistking.com) 

system. While not cheap, it is quiet. With 

the optional ZipDrip, any dripping from the 

nozzles can be eliminated. The starter kit 

from MistKing comes with a better timer, 

allowing timing down to one second. If you 

intend to get an integrated control system 

(see below) then purchasing à la carte may 

be better. The pump takes about five seconds 

to start up, so this time has to be added to the 

misting intervals. 

For the nozzles, I selected the straight 

type because they are the least conspicuous. 

Some people may want to install a canopy 

over the terrarium, and then the premium line 

may be superior if an acrylic lid is installed, 

which can be drilled by any horticulturist. 

I installed six misting nozzles about 9 

inches (22.5 cm) apart with black zip ties 

(cable ties affixed through small holes drilled 

in the black plastic rim on which the tank’s 

lid rests could also be used). The locking 

mechanism is pointing downward so the 

lid still closes reasonably tightly. The spray 

nozzles are almost completely hidden behind 

the top trim. 

The MistKing tubing is the same as that 

[2] Dendrobium christyanum (dwarf form) 

produces 1½-inch (3.75-cm) flowers 

that are larger than the pseudobulbs. It 

requires a summer rest, and is placed in 

a drier corner of the terrarium. Grower all 

plants shown: Daniel L. Geiger. 

[3] Dendrobium cucumerinum requires excellent 

air flow so it is placed next to one 

of the computer fans. 

[4] Specklinia brighamii produces yellow 

flowers measuring about 3 ⁄8 inch (1 cm). It 

grows better mounted on cork than potted 

and tolerates a lot of light (1,500– 

2,500 foot-candles) for a pleurothallid. 

for ¼-inch kitchen reverse-osmosis (RO) 

or water filtration systems. The tubing and 

connectors can be obtained at a hardware 

or specialized plumbing store, though usually 

not in black. Only the bulk-heads are 

difficult to find, i.e., the connectors to pass 

tubing through a wall. 

For the water supply, I opted for a 

second, small custom tank measuring 9½ 

inches tall × 8¾ inches wide × 24 inches 

deep (24 × 22 × 60 cm). It fits precisely next 

to the main tank behind the board with the 

controllers that I constructed. This second 

tank has one drilled hole with a T-junction 

connecting the supply line for the mister and 

the top-off system for the waterfall or swamp 

cooler. One can also connect the RO outlet 

directly to the mister or an intermediate tank 

with top-off system. 

Lighting Placing the growing 

lights on top of the tank makes removal of 

spent flowers or the application of fertilizer 

tedious. I opted for a hanging 60-inch (150- 

cm) Sunlight Supply Tek light with four T5 

80 W Giesemann Powerchrome Midday 

6000 K lamps with good color rendition 

index (CRI) of 90 1 . The whole fixture can 

easily be elevated or lowered to adjust light 

intensity and heat that affect the tank. 

I chose the four-bulb setup because it 

is the same width as the tank, and as sets of 

two bulbs can be turned on and off separately 

I can mimic early morning–late afternoon 

with only two bulbs turned on, and midday 

with all four. In addition, the heat output of 

the lamps is reduced. 

Two small computer fans on the fixture 

further mitigate the heating by the lamp. 

The temperature difference of the top glass 

with and without fans running is about 10 

F. Computer fans vary in terms of size, flow 

rate and noise levels given in decibels (db). 

As my tank is installed in our kitchen–dining 

area, I selected low-noise SilenX iXtrema 

fans (60 mm, 8 db). 

Waterfall A waterfall was to provide 

the heat-exchange mechanism for the 

swamp cooler (see below). I wanted a film 

of water running over a smooth natural rock 

to minimize splash. Marble can be cut thin so 

that the tank glass would not crack from its 

weight. A U-shaped aluminum profile was 

glued to the back of the rock. To make the 

water flow over the front of the rock face, 

the back trough has to be about 2–3 mm 

higher than the level edge of the rock due to 

the surface tension of the water. A Rio 180 

1 

Kindly communicated via email by Giesemann 

Lichtechnik & Aquaristic GmBH, Nettetal, Germany, 

which is good for T5HO lamps. Select full 

spectrum T5HO lamps have a CRI of up to 93. 

Regular fluorescent lamps such as cool white have 

a CRI of around 60. 

34 Orchids JANUARY 2012 www.AOS.org

mini powerhead pumping 45 gallons (170 

L) per hour at a height (= “head” in pump 

language) of 24 inches (60 cm) turned out to 

provide just about the minimal flow required 

(~8 gallons [30 L] per hour per inch of the 

width of the waterfall). 

Water loss due to splash can be avoided 

with the reservoir wall 1–2 inches (2.5–5 

cm) above the water level. If the rock for 

the swamp cooler is immersed at the bottom 

in the reservoir, it reduces splash and makes 

for a more quiet operation; if you like a little 

gurgling fountain, then let the water drip 

back down into the reservoir. 

Swamp cooler The tank can get 

excessively hot during peak summer days. 

Frozen plastic water bottles are an eyesore 

and can cause freezer burn on plants through 

direct contact. Some people have jerryrigged 

a small air conditioner with duct 

tape and hose, but that was not appealing to 

me. Additionally, air conditioners remove 

moisture from the air (hence, the dripping 

from AC units), and tropical orchids rather 

like moist environments. Accordingly, some 

indirect means of cooling had to be accomplished, 

which will not lower humidity in 

the terrarium. 

The idea is to cool a small volume of 

water quite strongly, and let that film of cold 

water flow over a flat piece of rock. A fan 

blows air over that wet rock transferring the 

heat of the air to the chilled water. Several 

coolers can be considered. Modified water 

coolers are unappealing and unproven. A 

pure electric cooler (Peltier elements), the 

Coolworks IceProbe, was too weak and 

failed after one day of operation. 

The next step up in cooling power is 

a more conventional flow-through chiller 

available in the aquarium trade. After much 

deliberation and with the help of staff at a local 

aquarium shop I settled on a JBJ Arctica 

1/15 HP chiller. It can cool the tank from 85 

F (29 C) to 76 F (24 C) with water at 60 F in 

about an hour, the compressor active about 

half the time. There is a temperature gradi- 

5 6 

[5] Dendrobium aratriferum grows in a small 

clay pot. It produces flowers that are half 

the size of the pseudobulb plus leaf. The 

flowers last only a single day, but are 

spectacular. 

[6] Schoenorchis fragrans is a miniature — 

it grows only 1 inch (2.5 cm) tall. A more 

recent experiment, it seems to do well in 

the terrarium at intermediate light levels. 

[7] Leptotes bicolor is a rather large plant for 

a terrarium; its thick pencil-shaped leaves 

are about 3–4 inches (7.5–10 cm) long. 

It produces rather long-lived and quite 

large flowers. 

7 

www.AOS.org JANUARY 2012 Orchids 35

8 

Behind the Curtain 

[8] Epidendrum peperomia has overgrown 

its original mount and is covering some 

of the ghost wood branches. It flowers 

from about December through February 

or March. 

[9] Dendrobium pachyphyllum has redbrown 

pseudobulbs that contrast with the 

green leaves. It grows on a stick-mount. 

The ¼-inch (.6-cm) flowers last only for 

one or two days. 

The automated control for the terrarium’s 

temperature and humidity is based on 

the Digital Aquatics HerpKeeper system 

and is mounted on a wood board with all 

components and wires labeled. 

1. Main powerstrip connected to GFI 

protected outlet. 

2. HerpKeeper control unit. 

3. HerpKeeper controlled powerstrip 

with indicator lights. 

4. Hub to connect additional units, 

essentially a small Ethernet hub, using 

ethernet cables. 

5. Moonlight controller. 

6. Three variable power transformers for 

fans. 

7. Mistking pump. 

8. Mistking zip-drip. 

9. Mistking power supply. 

10. Main power cord. 

11. White polyethylene tube connecting 

water reservoir behind control board 

to automatic top-off of reservoir for 

waterfall. 

12. Black polyethylene tube supplying 

spray nozzles. 

13. Under-tank heat element power cords. 

— Daniel L. Geiger. 

9 

ent of about 2–4 F along the 5-foot (1.5-m) 

terrarium. 

The plumbing is accomplished with 

various flexible hoses and some hard sprinkler 

tubing. The installed baffle at the entry 

to the reservoir turned out to be unnecessary 

with the 110 gallons (416 L)/h pump and the 

rather wide 1-inch (2.5-cm) tube diameter. 

The chiller can be detached to be stored 

elsewhere, and the tubing is out of sight. 

The chiller is only hooked up for about three 

months of the year. 

Auto top-off (ATO) Some of the 

water in the swamp cooler will be gradually 

lost due to evaporation and some splash. The 

pump and the chiller should not run dry, so 

an ATO is included. It consists of distilled 

water line fed by gravity flow, and a simple 

float valve. Gravity feed works by a higher 

water level in the reservoir tank, and no 

active pump is associated with it. I prefer 

gravity feed from a relatively small reservoir 

as opposed to having the water supply 

plumbed into a RO system, because in case 

of a malfunction, the tank can only flood to 

a relatively small extent. In summer about 

one liter per week is lost from the reservoir. 

In winter with the water being heated, the 

evaporation rate is much greater, and the 

ATO becomes convenient. 

Heating In winter the terrarium has 

to be heated at night for it to stay above 60 F 

(16 C). I installed three under-tank heat pads 

measuring 11 × 17 inches (27.5 × 42.5 cm), 

25 W each, covering most of the underside 

of the tank. Heat pads come in two strengths: 

desert and rainforest, the latter reaching 

about half maximum temperature. As the 

roots of the plants should not be cooked I 

chose the latter (www.exo-terra.com/en/ 

products/heat_wave_rainforest.php). However, 

they did not produce enough heat. How 

else to heat the tank? If it is possible to cool 

the tank indirectly through the waterfall, 

then it should also be possible to heat by the 

same principle. A secondary benefit is higher 

humidity, as the vapor pressure of water 

increases with temperature. A small 100 W 

aquarium heater (www.planetrenadirect. 

com/category/planetrena.rena.smartheater/) 

set to 80 F (27 C) kept the tank above 60 F 

(16 C) throughout the night. Humidity did 

increase and less misting was required, but 

the waterfall compartment lost water at a 

quicker rate. Given that the water heater is 

much cheaper ($25) compared with the three 

heat pads ($100) and produces better results, 

the water heater may be a good option for 

budget conscientious people. Whether some 

warming of the root zone has special merits 

on its own remains to be investigated. 

Before I had the ATO installed, I manually 

topped off the waterfall reservoir with 


filtered tap water. After about two months, 

I cleaned the waterfall and finished the ATO 

setup. I noticed strong mineral deposits on 

the heater. With hindsight, this is not surprising, 

as I added mineral-rich tapwater, and 

only distilled water evaporated, leading to 

an accumulation of minerals. Using distilled, 

deionized, or RO water can help to reduce 

the problem. 

Antifogging ventilation The 

front glass should be mostly free of water 

droplets. The misters will deposit some 

spray on the front glass, and under high 

humidity conditions, water may condense 

on the front glass obstructing the view. This 

problem had been addressed by a frog keeper 

with an ingenious venting system for a 

multitank arrangement (See www.mistking. 

com/pages.php?page id=7). Unfortunately, 

my attempts at replicating this approach 

failed. After some experimentation, four 60 

mm downdraft fans, wired sequentially to 

one transformer and 1 foot (30 cm) apart, are 

affixed to the top rim of the terrarium with 

cable ties. While not 100 percent effective, it 

does improve the viewing conditions. 

Controls for temperature 

and humidity The control of the 

swamp cooler, heating elements and misters 

to maintain temperature and humidity within 

desired ranges was insufficiently developed. 

When it comes to advanced habitat control, 

the coral reef keepers are the clear leader 

in the field. Digital Aquatics has recently 

released the first such controller specifically 

designed for terrariums, the HerpKeeper. 

After having tried numerous timers that 

are bulky and often cease to work after a 

while, the HerpKeeper is the answer to all 

those previous problems. I can only provide 

a brief overview here. The controller box is 

similar to a home thermostat with a few more 

options. An attached temperature and humidity 

probe gives its readings to the central controller. 

The controller can use timer functions 

to turn on/off outlets on modified four-outlet 

power strips at set times, intervals or when 

certain temperature or humidity conditions 

are reported from the probe. The temperature 

and humidity readings can also enact overrides 

(so-called alarms). For instance, I have 

set the mister to mist from 8 am to 8 pm for 

15 seconds every hour, unless the humidity 

is greater than 70 percent, but between 8 pm 

and 8 am the interval is three hours. 

While the product’s capability and 

modular expansion options are staggering, 

the documentation is rudimentary. There is 

an active bulletin-board forum and email 

technical support is exquisite. 

Controlling clutter The new 

orchid terrarium has a bit more complex infrastructure. 

However, I wanted to hide any 

10 

cables and equipment as much as possible 

as not to distract from the intrinsic beauty of 

the display and the plants. As I do not have 

space underneath the bench, I use a wood 

board in front of the water tank to mount the 

HerpKeeper controller, power strips and the 

MistKing pump. The various plugs, outlets 

and tubes can be quite confusing, so I used 

a label maker to identify each, some even in 

more than one place. Cable ties additionally 

help to keep things neat and in place. 

Landscaping Landscaping involves 

several steps. First any localized 

pressure from pots and branches should be 

distributed so that the bottom glass does 

not break. At the bottom. I placed sheets of 

¼-inch (0.6 cm) Styrofoam recycled from 

packaging material. 

In the old tank, I used orchid bark and 

rocks for landscaping. The orchid bark does 

decay with time, so I went with expanded 

11 

12 

[10] Notylia barkeri, one of the largest plants 

in the terrarium, bears 1-foot- (30-cm-) 

long inflorescences with more than 100 

fragrant flowers. 

[11] Mystacidium capense produces inflorescences 

with geometrically arranged 

spurs. It flowers occasionally in the 

terrarium. 

[12] Specklinia grobyi is a free-flowering 

pleurothallid that can also take a lot of 

light. The inflorescences are about 

3 inches (7.5 cm) long and bear 3 ⁄8-inch 

(1-cm) flowers that last about a week. 

clay often used in hydroponics. It absorbs 

some water and thus, provides a humidity 

buffer but does not decompose. To hide 

pots at the front glass, I used thin pieces of 

slate, and where I wanted to create a bit of 


space between the plant and the front glass, 

I placed some cork used to mount epiphytic 

species. 

For the epiphytes, I first transferred the 

old branches setup. Some of the orchids had 

grown onto those logs, so best to keep them 

where they are thriving. I added several 

additional pieces that also hide electrical 

cables. Some space was left intentionally 

open to allow for some taller species, such 

as Lockhartia lunifera. 

Pest control In general, I don’t 

mind a few animals in the tank. But vermin 

munching on buds or flower cannot be tolerated. 

Some isopods (roly-polies) of unknown 

source started to nibble on orchid buds. Dismantling 

the tank was not an option, manual 

eradication was bound to fail and I was not 

keen on using some strong chemical insecticide. 

On Orchidboard (www.orchidboard. 

com/community/terrarium-gardening/109- 

must-read-getting-rid-pests-terrariums. 

html) someone suggested dry ice (solid 

carbon dioxide) to suffocate the intruders. 

Some grocery stores, beverage or party 

supply houses sell dry ice for about US$1 

per pound; 1–2 pounds should be more than 

enough for a 50–100 gallon (189–378 L) 

tank. Place the dry ice into a plastic container 

and wedge it under the tank cover, then add 

some hot water to sublimate the solid dry ice 

into gaseous CO 2 

. The CO 2 

fills the tank with 

a thick fog, and the isopods start to die off 

immediately. An hour later, the dry ice had 

vanished and the tank was clear of the little 

miscreants. Plants did not show any adverse 

effects as the temperature drop in the tank 

was only about 5 F, but be careful that the 

dry ice container does not touch any plants as 

it will cause freezer burn. The treatment had 

to be repeated once to get rid of the newly 

hatched individuals; the CO 2 

treatment does 

not kill eggs. Only use dry ice in well-ventilated 

areas as it is a breathing hazard. 

WEEDS A second problem is unwanted 

plant growth. Moss overgrew Acianthera 

luteola 2 and suffocated the plant eventually. 

On the other hand, Bulbophyllum acutebracteatum 

took advantage of the moist grounds 

to root. The mosses also partially covered 

Pinalia 3 amica and Dendrobium rigidum 4 . 

The only remedy has been manual removal 

with fine forceps, but it is almost impossible 

to remove all pieces, so the procedure needs 

to be carried out periodically (every four to 

six months). 

PLANTS How are the plants doing? 

Most of the original plants have been 

surviving; the losses were mainly the more 

cold–intermediate instead of warm-growing 

species (e.g., Masdevallia spp., Lepanthes 

calodictyon). The best species are Angraecum 

distichum, Bulbophyllum roxburghii, 5 

Bulbophyllum acutebracteatum, Dendrobium 

lichenastrum, 6 Den. rigidum, Epidendrum 

peperomia, 7 Lockhartia lunifera, Specklinia 8 

tribuloides, Specklinia 8 grobyi and Cattleya 9 

cernua, disregarding the nibbling of the buds 

by the isopods on the last. 

Ceratocentron fesselii is fickle. It set 

several inflorescences that started to mature 

over months, but then all of them wilted 

and fell off. Its leaves also shriveled, which 

leads me to believe, that I may have kept it 

in too bright light (1,150 foot-candles, shade 

recommended [http://www.orchidspecies. 

com/ceratfesseli.htm.]). Remounting it 

with moss and positioning it in shade shows 

some recovery, but it has not flowered yet. 

Lockhartia lunifera can grow 18 inches (45 

cm) tall and requires an appropriately sized 

space. Pinalia amica is a slow grower, but 

the current five-pseudobulb plant produced 

three new ones. The rather short-lived inflorescences 

are spectacular, and worth the 


wait. Platystele ortiziana is almost dying 

in summer, and usually recovering in winter. 

I received the plant potted in peat moss 

and have kept it that way. I suspect, excess 

water may also be a problem, and I have 

seen it recommended to be stick-mounted 

(http://www.andysorchids.com/pictureframe. 

asp?pic=images/Species/6310med.jpg&Pi 

cId=6310&PicNam=Platystele-ortiziana.). 

Similar overwatering problems may also be 

responsible for leaf loss in Podangis dactylo- 

ceras. I received it in orchid bark and perlite, 

but dry-out between watering is recommended 

(www.orchidspecies.com/podangisdactyleceras.htm.). 

I repotted it in expanded clay 

pellets but it did not survive. Masdevallia 

Pixie Shadow (infracta × schroederiana), an 

intermediate grower, has excellent vegetative 

growth in a shady portion, but did not produce 

flowers in the terrarium. In the greenhouse, 

at low to intermediate light levels, it is now 

producing a flower bud. 

[13] Bulbophyllum roxburghii produces 

2-inch (5-cm) umbrella inflorescences. 

2 

Formerly Pleurothallis caespitosa. 

3 

Formerly Eria. 

4 

Formerly Dockrillia rigida. 

5 

Formerly Cirrhopetalum sikkimense. 

6 

Formerly Dendrobium prenticei. 

7 

Formerly Epidendrum porpax. 

8 

Formerly Pleurothallis. 

9 

Formerly Sophronitis. 

10 

Formerly Haraella odorata. 

11 

Formerly Oncidium variegatum. 

13 


14 15 

Some odd failures include Haraella 

retrocalla, 10 a well-regarded terrarium species 

and exquisite performer in my setting 

for several years, which eventually started to 

lose one leaf after the other, until none were 

left (pathogen?). A new specimen started to 

show the same symptoms, and has now been 

moved to a shaded greenhouse. Tolumnia 

variegata 11 produced one spike and multiple 

new fans, but the plant’s leaves yellowed 

and eventually died, which may be a sign 

of too much light. 2,750–3,750 foot-candles 

are recommended, but even cutting that in 

half (see below for rationale) may still be 

too much. Masdevallia zahlbruckneri initially 

showed good growth and flowered, 

but after a few months dropped leaves and 

aborted flowers. It now lives in the shady 

greenhouse. Bulbophyllum longissimum is a 

very large species for a terrarium with leaves 

about one foot long. The flowers are just too 

gorgeous to pass up. After keeping the plant 

in the terrarium for about nine months with 

good vegetative growth, it has now been 

moved to the greenhouse, where it produced 

multiple inflorescences. 

Some species are more recent additions, 

and while the initial observations are promising, 

it is too early to tell whether they will 

work in the long run: Isabelia 12 violacea 

(flowers, new pseudobulbs), Cischweinfia 

sheehaniae, Bulbophyllum lasiochilum 

(fickle, but if well grown then with excellent 

growth, several flowers), Neofinetia 

falcata and Tolumnia calochila, 13 Restrepia 

muscifera (flowered, producing new leaves), 

12 

Formerly Sophronitella. 

13 

Formerly Onchidium calochilum. 

14 

Formerly Cadetia. 

15 

Formerly Diplocaulobium. 

16 

Formerly Pleurothallis caespitosa. 

Dendrobium 14 taylori (flowered, good 

growth), Dendrobium 15 aratiferum (new 

leaves), Dendrobium 8 cucumerina (flowered, 

new leaves), Masdevallia wendlandiana 

(flowered, good growth). Dendrobium 

christyanum (dwarf form) has produced 

flowers since being placed in the terrarium, 

but has not yet produced new bulbs. 

On pages 44–45, I list additional intermediate-warm-hot 

species that have been 

recommended for terrarium culture. This is 

a collation from internet boards, personal 

experience, and personal recommendations. 

I provide species name, temperature range 

(winter night minimum, summer daytime 

maximum), recommended published peak 

natural light intensity (see below for discussion 

and caveats), flowering season and 

watering requirements. Some large species 

have been omitted for practical reasons (e.g., 

Vanilla planifolia). Entries followed by an 

asterisk (*) are those that have worked well 

for me or that are recommended by multiple 

independent sources. 

Light levels for plants In 

general, published light level indications 

seem to be on the high side when applied to a 

terrarium setting. Specifically, Epi. peperomia 

has suggested light levels of 2,500–3,500 footcandles 

(www.andysorchids.com.) and bright 

to partial full sun (www.orchidspecies.com/ 

epiporpax.html), while my plant is showing 

visible signs of burning at 1,300 foot-candles 

(yellow leaves with brown edges), with 

shaded portions (250–500 foot-candles) being 

much healthier (solid green leaves). Note that 

the recommended and measured light intensities 

differ by an order of magnitude. Some 

plants have good vegetative growth in too low 

light but do not flower. This possibility can be 

excluded as my specimen flowers profusely 

in fall through spring. 

[14] Mats of moss growing on orchid bark. It 

suffocated Acianthera luteola 16 (brown 

leaf), but provided good rooting rounds 

for Bulbophyllum acutibracteatum. The 

spread has to be controlled manually. 

[15] An isopod feeds on the leaves and buds 

of orchids. The population was eradicated 

with dry ice. 

This mismatch may be related to peak 

intensity compared to overall mean intensity; 

I presume that peak intensities are given in 

the cited sources. Outdoor light intensity 

is weaker by a factor of four to five in the 

early morning and late afternoon, compared 

with the peak hours (http://naturalfrequency. 

com/articles%5Caveragehourly), and varies 

by a factor two to three over the course of 

the year (www.solarpanelsplus.com/solarinsolation-levels/) 

at intermediate latitudes. 

In a terrarium, lighting is mostly provided 

by a constant, artificial light source, usually 

on a 12 hours on, 12 hours off cycle at full 

output. The total amount of light received 

over the course of a day is possibly a better 

measure, than the peak intensity. It is possible 

to convert the amount of natural daylight to 

the amount of artificially provided light in 

a terrarium setting (see the AOS website, 

www.aos.org, for mathematical derivation). 

Based on those calculations, around half of 

the usually indicated peak intensities are appropriate 

under continuous terrarium lighting 

regimes. We need to realize, however, that a 

number of factors are ignored here, including 

the spectrum of the light sources, spectral 

response of light meter, incident angle of light, 

nonlinear response of plants to light intensity 

(threshold illumination to photoinhibition), 

adaptation of plants to various light conditions, 

species-specific factors, population and 


strain-specific requirements and inaccuracy 

of published information. Hence, both the 

published recommended light levels as well 

as the correction factor advocated here should 

be taken as a starting point for experimentation 

and not a definitive value. 

For instance, a strong peak in the T5HO 

lamps at approximately 430 nm (www. 

giesemann.de/63,2,,.html.) coincides with 

a high relative photosynthetic rate of plants 

at that wave length (www.life.illinois.edu/ 

govindjee/paper/fig5.gif.). Natural sunlight 

at 430 nm has approximately 80 percent of 

maximum intensity (at approx. 480 nm [e.g., 

http://www.imaging-resource.com/ARTS/ 

TESTS/SUNSOURCE2.HTM]), whereas 

the T5HO lamps emit about twice as much 

light at 430 nm compared with average 

maximum. 

I measured light intensity given off by 

the T5HO lamps. Right at the lamp approximately 

4,000 foot-candles are produced. At 

a sensible closest distance of about 5 inches 

(12.5 cm), which prevent excessive heat to 

reach the plants, and allows me to open the 

top of the terrarium, about 1,300 foot-candles 

are given off with two bulbs, about 3,000 

foot-candles with all four running. These 

values drop to 620 foot-candles and 1,200 

foot-candles in the middle of the terrarium, 

and 250 foot-candles and 500 foot-candles at 

the bottom of the terrarium, respectively. 

If we assume as a first approximation 

that suggested light peak intensities should 

be cut in half for terrarium culture, then the 

maximum light intensity reaching any plant 

at any time should be around 1,500–2,000 

foot-candles. These values are reached with 

the T5HO fixture at a distance of approximately 

8–10 inches (20–25 cm) from the top 

of the terrarium. They will depend on many 

variables, such as exact bulb type, reflector 

design, and age of bulb. 

Ideally, the daily and seasonal variation 

is mimicked in the terrarium. My four bulbs 

can be turned on and off in pairs. Two bulbs 

are on for 12 hours, and I only run all four 

bulbs during four hours. With this approach 

I can only double or halve the light intensity 

during the day, still falling short by a factor 

of two compared to natural daily light 

intensity changes. 

In the aquarium trade there exist dimmable 

ballasts for metal discharge lamps, 

which can be managed by external controllers. 

Some metal discharge lamps have a 

excellent color spectrum (e.g., Iwasaki EYE 

colorArc with excellent CRI 96, and color 

temperatures of 4500 or 6500 K). Hence, 

there may be ways to deliver more natural 

light changes over the course of a day and 

17 

Formerly Bulbophyllum tingabarinum. 

throughout the seasons. For my application, 

the esthetics are better served by T5 fixtures 

than by metal discharge lamps. EcoZone 

(http://ecozonevivarium.com) lists some 

dimmable T8 and T5 fixtures. 

The latest developments are LED light 

sources. They are used by some coral reef 

aquarists, and they also need high light output, 

but have different spectral requirements 

(higher UV contribution). LEDs use much 

less energy for the same light output, and 

also produce less heat. This area is promising 

in the future. 

COST What does all this add up to? 

Below is a breakdown of costs: 

♦ 100 gallon (378 L) custom tank 

($700) 

♦ Lights ($500) 

♦ Misting system ($400) 

♦ Chiller ($500) 

♦ Waterfall ($50) 

♦ Heating pads ($100) 

♦ Water heater ($25) 

♦ Controller (base: $200; with some 

optional additions $400) 

♦ Miscellaneous ($300) 

♦ Total (approximately $3,000, without 

plants). 

While not cheap, the final result is both 

pleasing to the eye as well as to the plants. 

For me it is well worth it. 

References 

Geiger, Daniel L. 2008. Orchids Under Glass: Growing 

and Flowering Plants in a Terrarium. Orchids 

77(11):846–853. 

Daniel L. Geiger, PhD, is a marine invertebrate 

systematist working at the Santa 

Barbara Museum of Natural History. He is a 

member of the Orchid Society of Santa Barbara 

and the California Native Plant Society, 

and has been growing native California 

orchids outdoors, and more recently added 

an indoor terrarium with tropical orchids. 

Santa Barbara Museum of Natural History 

– Invertebrate Zoology, 2553 Puesta del 

Sol Road, Santa Barbara, California 93105 

(email Geiger@vetigastropoda.com). 

[16] The rather uncommon orange-flowered 

Bulbophyllum pecten-veneris 17 is a 

compact plant with about 2-inch- (5-cm-) 

long inflorescences. 

[17] Masdevallia zahlbruckneri is one of the 

masdevallias that tolerates somewhat 

higher temperatures, but requires low 

light (less than 500 foot-candles). 

[18] Bulbophyllum newportii is a new addition 

to the terrarium. The straight 

inflorescence is about 1½ inches (3.7 

cm) tall and bears about five to seven 

¼-inch (0.6-cm) flowers. 

16 

17 

18 


daniel l. geiger 

A Selection of Orchids 

Orchid Temperature Light Flowering watering 

(F) 

(foot-candles) 

Aerangis luteoalba var. rhodosticta warm light shade winter spring moist 

Amesiella minor 52–80 500–1,500 winter moist 

Angraecum distichum* 500–1,000 all year 

Angraecum didieri 55–58 1,500–2,500 all year moist 

Ascocentrum 58–88 2,500–3,500 spring moist 

Asconopsis 

(Ascocentrum × Phalaenopsis) 

Bulbophyllym 

lasiochilum 

Cattleya cernua 


Bulbophyllum acutibracteatum* warm–hot 500–1,000 autumn moist 

Bulbophyllum alagense 55–85 500–1,500 all year moist 

Bulbophyllum clandestinum 1 cool–warm partial shade spring–autumn 

Bulbophyllum corolliferum 2 58–88 1,200–2,500 

Bulbophyllum fascinator 58–88 1,500–2,500 summer–autumn moist 

Bulbophyllum inunctum cool–hot light shade summer–autumn moist 

Bulbophyllum lasiochilum* 58–88 2,500–3,500 all year moist 

Bulbophyllum longissimum 58–88 1,500–2,000 all year moist 

Bulbophyllum odoratissimum cool–warm partial shade spring–summer moist 

Bulbophyllum pardalotum hot shade all year 

Bulbophyllum pectin-veneris 3 60–90 2,000–3,000 spring–summer moist 

Bulbophyllum roxburghii* hot 500–1,000 winter–spring moderate 

Cattleya 4 cernua* 55–85 2500–3,500 winter–spring moist 

Cattleya 5 lundii 45–98 1,500–2,500 winter–spring moist 

Ceratocentron fesselii* cool–hot shade winter moist 

Ceratostylis philippinensis 55–85 1,500–2,500 winter–spring moist 

Chiloschista viridiflava 58–88 2,500–3,500 variable moist 

Christensonella 6 uncata 58–88 1,500–2,500 all year moist 

Cischweinfia pusilla 58–88 1,500–2,500 all year moist 

Dendrobium 

christyanum 

Dendrobium taylorii 



Dendrobium aberrans 55–85 2,500–3,500 winter–spring moist 

Dendrobium 7 aratriferum 55–85 2,500–3,500 all year moist 

Dendrobium atroviolaceum 58–88 2,500–3,000 winter–spring moist 

Dendrobium christyanum* 40–95 2,500–3,500 spring–summer moist 

Dendrobium 7 chrysotropsis 

cool–intermediate 

Dendrobium 8 cucumerina* cool–hot part shade winter– spring moist 

Dendrobium kingianum 40–95 2,750–3,750 winter–spring moist 

Dendrobium lichenastrum 9 * 58–85 2,500–3,500 all year moist 

Dendrobium 8 linguiforme 55–58 2,750–3,750 (autumn) winter (spring) 

Dendrobium oligophyllum 55–85 2,500–3,500 autumn–winter moist 

Dendrobium 8 rigidum* 45–98 500–3,500 all year moist 

Dendrobium taylorii 10 * 55–85 500–1,500 all year moist 

Dendrobium toressae cool–hot part sun spring 

Dendrochilum wenzelii 55–85 2,000–3,000 spring–summer moist 

Dinema 11 polybulbon 45–98 2,500–3,500 autumn–winter moist 

Dyakia hendersoniana hot part shade spring summer moist 

Epidendrum longirepens 58–88 1,500–2,500 all year moist 

Epidendrum nocturnum 58–88 500–1,500 all year? moist 

Epidendrum peperomia 12 * 55–85 2,500–3,500 winter–spring moist 

Epidendrum schlechterianum 13 warm–hot shade winter–spring 

Haraella retrocalla 


Gastrochilus matsuran intermediate–warm intermediate moist 

Gomesa 14 croesus cool–hot part shade spring–summer moist 

Gomesa 15 eleuterosepala† cool–hot(?) part sun winter 

Gomesa 16 radicans 55–85 2,000–3,000 summer–autumn 

Grandiphyllum auricula 17 55–85 2,500–3,500 

Grandiphyllum 14 edwallii 55–85 2,500–3,500 autumn–winter 

KEY † http://www.orchidspecies.com/rodrieuletherosepala.htm lists the species as cool growing, whereas the vendor (Santa Barbara 

Orchid Estate) labels it as temperature tolerant. * recommended by multiple sources or growing well in the author’s terrarium. 

Temperature Cool: ~52–75 F; Intermediate: ~58–80 F; Warm: ~60–85 F; Hot: ~70–95 F. Watering Moist: no prolonged drying 

of roots, watering usually once a day; Moderate; some drying of roots between waterings, watering usually every two to three days. 

Light Shade: 500–1,500 foot-candles (f-c); Part Shade: 1,500–2,500 f-c; Intermediate: 1,500–3,500 f-c; Part Sun: 2,500–3,500 f-c. 


Grown in the Terrarium 

Orchid Temperature Light Flowering watering 

(F) 

(foot-candles) 

Haraella retrocalla* 55–85 500–1,500 all year moist 

Isabelia 18 violacea* 55–85 2500–3,500 winter–spring moist 

Leochilus carinatus cool–hot light shade spring–winter moist 

Leptotes pauloensis* cool-hot light shade spring–winter moist 

Lockhartia lunifera* 58–88 1,500–2,500 all year (?) moist 

Macodes petola warm–hot part shade autumn 

Malaxis sp. 58–90 500–1,500 seasonal moist 

Masdevallia floribunda 45–98 500–1,500 summer 

Masdevallia herradurae 55–85 500–1,500 summer–autumn moist 

Masdevallia Pixie Shadow intermediate shade spring–summer 

(infracta × schroederiana) 

Masdevallia wendlandiana 65–95 500–1,500 all year moist 

Masdevallia zahlbruckneri cool–hot 500–1,500 all year moist 

Neobathiea grandidierana cool–warm shade spring 

Neofinetia falcata* 55–85 2,500–3,500 all year moist 

Nephelaphyllum sp. warm shade spring summer 

Notylia barkeri* intermediate–hot 1,000–2,000 winter–spring moist 

Pinalia 19 amica* cool–warm part shade spring moist 

Platystele ortiziana* 55–85 500–1,500 all year moist 

Pleurothallis allenii 55–85 1,500–2,500 spring–autumn moist 

Pleurothallis fastidiosa warm–hot shade spring 

Pleurothallis luctuosa 55–85 1,500–2,500 all year moist 

Podangis dactyloceras 58–88 2,000–3,000 spring–summer–autumn moist 



Isabelia violacea 

Lockhartia lunifera 

Restrepia muscifera* 55–85 500–1,500 all year moist 

Robiquetia sp. cool–hot part sun moist 

Sarcochilus ceciliae cool–warm part sun spring–summer moist 

Scaphosepalum fimbriatum warm–hot 500–1,000 all year 

Specklinia 20 brighamii* 58–88 1,500–2,500 all year moist 

Specklinia 20 grobyi* 40–95 1,500–2,500 all year moist 

Specklinia 20 megalops 55–85 1,500–2,500 all year moist 

Specklinia 20 microphylla warm–hot part sun summer–autumn 

Specklinia 20 tribuloides* 55–88 500–1,500 all year 

Stelis microchila 21 warm–hot 500–1,500 spring–autumn 

Stelis morganii cool–warm shade spring 

Stelis 20 restrepioides 52–80 500–1,500 winter– spring 

Tolumnia calochila 22 * 65–95 2,500–3,500 all year moist 

Tolumnia variegata 23 * 60–90 2,750–3,750 winter–spring 

Trichoglottis triflora 45–98 2,000–3,000 winter–spring moist 

Zootrophion atropurpureum warm 500–1,000 summer–winter 

Zygostates lunata warm shade autumn–winter 

Zygostates pellucida* warm–hot shade variable moist 



Restrepia muscifera 

Tolumnia calochila 

1 

Formerly sessile. 

2 

Formerly Cirrhopetalum curtsii. 

3 

Formerly Cirrhopetalum tingabarinum. 

4 

Formerly Sophronitis. 

5 

Formerly Laelia. 

6 

Formerly Maxillaria. 

7 

Formerly Diplocaulobium. 

8 

Formerly Dockrillia. 

9 

Formerly prenticei. 

10 

Formerly Cadetia. 

11 

Formerly Encyclia. 

12 

Formerly porpax. 

13 

Formerly congestoides. 

14 

Formerly Oncidium. 

15 

Formerly Rodrigueziopsis. 

16 

Formerly Ornithophora. 

17 

Formerly Oncidium harrisonianum. 

18 

Formerly Sophronitella. 

19 

Formerly Eria. 

20 

Formerly Pleurothallis. 

21 

Formerly barbata. 

22 

Formerly Oncidium calochilum. 

23 

Formerly Oncidium variegatum. 

— Compiled by Daniel L. Geiger. 


Zygostates pellucida

Orchids Under Glass - Daniel Geiger's Homepage

Create successful ePaper yourself

Delete template?

Save as template?