Orbs are the most debated part of the evidence at this point in the history of paranormal investigation and although dust, water droplets, snow, pollen, water vapor, mold spores and other such material can mimic the way orbs look physically, there are important differences that must be carefully examined and identified before an anomaly can truly be called an orb. As a generally accepted rule there are three questions an investigator can use to determine if an orb is genuine or if it is some sort of airborne particulate matter.  They are as follows:

Is the orb in question perfectly circular?  By definition an orb is a circular or sphere shaped object. If an orb in question is elliptical, oval, diamond shaped, etc. it is not a true orb...therefore, the potential for an airborne particle is almost 90%

Is the orb three dimensional?  Does it stand out from its surroundings and/or have depth in the photograph?  If not it is probably airborne particulate matter or something on the camera lens.

Is the orb self-luminous - does it emit its own light?

Orb Brightness:

HDO – High Density Orbs are the least common of the two types of orbs. Generally, these orbs are the only anomaly in a photo, are opaque (solid) in appearance and glow brightly from within. These orbs are usually of magnitudes two (bright) and three (super bright).

LDO – Low Density Orbs are the more common of the two types of orbs and are generally what is thought of when the subject of orbs is mentioned as these can very closely resemble airborne particulate matter. The primary difference between these orbs and airborne matter is that, although they are translucent, they are light emissive. This type of orb is of a magnitude zero (almost completely dark) or one (dim).

Much like stars in the night sky, orbs can be measured by their magnitude (brightness) in a photo. Although this is really only a theoretical system for classifying photographic data for investigation purposes I believe that it will prove useful for those serious enough to explore it.  The basic idea behind orb magnitude is that investigators can now classify orbs so that they may be understood in relation to events such as solar flares, geomagnetic storms, etc. so that a pattern of activity may be established. Orbs rated by magnitude will follow an ascending scale from zero (0) least bright to three (3) most bright.

The Actual Theory on Orbs:
Refraction Rings

Arguably, the most common causes of orbs are dust and other airborne particles, or even small flying insects. When solid particles are
floating just inches in front of the camera's lens, then the photographic flash makes them very bright. This particle is out of focus, so it just shows up as a round spot of light in the picture. Moisture in the air can also catch the light and cause orbs. It is important to know if the air was damp the day the picture was taken or if it was raining or snowing. These particles, when exposed to the flash, because refraction rings that form within the orb which can easily be seen without much effort. These rings are caused by the particle refracting the light back to the lens; different particles may have different rings and different types of rings.

Orbs can also be created by bright lights in an area where the photo is being taken, by angles of light and by many types of artificial lighting. Lens flare is caused when there's a bright light, like the sun, in the picture or just outside of it. The source of light reflects off a portion of the lens, and creates internal reflections of the aperture of the camera. Colored orbs may be just dependent on what light refraction's were present when the photos were taken. While rare (except in some heavy swampy marsh areas), ball lightning may also be mistaken for an orb and may fit all the requirements previously stated.

Some have argued that these anomalies could still be dust or airborne particulate matter. Potentially, however unless the dust was radioactive it probably would not glow of its own accord. Dust, water vapor/droplets, pollen, mold spores and other such materials can only be photographed within a few feet of the camera which is generally the effective flash distance – any orb outside of this range that meets the three primary criteria could not be disproved by this line of reasoning (unless it was a bright light seen off in the distance.. such as building lights, lights off of towers, solar grave marker lights, reflections of reflective surfaces, or dim lights commonly seen within the home). More often than not I have seen photos taken by someone that is of some sort of airborne particulate matter which they believe is an orb. When I have tried to explain, by the reasoning utilized in the three questions previously outlined, that what they have photographed is not a real orb some people have become upset and insist that it is an orb that they photographed. Some people so desperately want to believe that they have captured ghostly activity in their photo and because of this they will not listen to any logic or scientific reason that would indicate anything other than what they want to believe. If you are evaluating a photograph and the anomaly in the photo does not meet the minimum three requirements as stated above, whatever is in the photo is, almost without question, not a true orb. In order to avoid this “is it or isn't it?” argument if someone still insists that it is an orb because it has an unusual color you may want to refer to the following list:

Yellow, Orange or Red orbs = Pollen
Green or Blue orbs = Mold Spores
Grey orbs = Dust
White or Blue orbs = Snow, water or dust

This Theory was formed out of discussions between Larry Eissler and Dan Norvell of Small Town Paranormal.