History of Delay:
Originally (as early as 1920) delay was obtained by magnetic recording devices that recorded the audio with one head and reproduced the audio with a different head. The spacing between the heads and the speed of the magnetic medium (the recording tape) determined the delay. Several units were made with one record head and several reproduce heads to give several possible delays simultaneously.
Adding feedback to a delay will generate repeated echoes.  Some early tape delay units provided a means of sending the delayed signal back into the input of the unit to provide a repeated echo effect.  The brand "Echoplex" was one such unit that was very popular in the 1960's. 
In the late 60's devices were marketed that had a small speaker work into a long tube with a microphone at the end of the tube. Such devices required extensive equalization to have a flat frequency response, tended to be noisy and only had fixed delay times.
In the 70's two electronic devices were marketed to achieve delay. The first such device was the analog delay line. The fist analog delay lines used filter circuits with a flat frequency response in the audio-frequency range. The filter slightly delayed the audio signal and longer delay times could be obtained by putting the signal through a series of filter circuits. Later analog delay lines sampled the audio several times a cycle, delayed the samples and then reconstructed the audio back into full waveforms. By analog we mean that the signal is a representation of the actual waveforms.
Also first appearing in the 1970's was the digital delay line. Almost all delay and reverberation devices used in the modern control room is based on the digital delay line. In the digital delay line the audio waveform is sampled and then converted to a digital signal (a series of pulses which define a number representing the audio level at a given instant). The digital signal is then delayed and converted back into an audio signal.
In the early and mid 1990's all digital mixers (such as the Yamaha 02R) and modular digital multitrack recorders (such as the ADAT) appeared and had as one of their features a channel or track delay (usually up to 100 ms.) on every channe, making the use of delay very common.  
The Basics of Reverberation:
The digital reverberation device is based on several digital delay lines. In the modern control room, most of the reverberation comes from the digital reverberation device. At a touch of a button or two, the engineer is able to obtain very realistic effects that simulate real acoustic environments. In addition there are a large variety of unreal effects that have been created and are popularly used.
The delay devices have various "programs" which determine how many delay lines are used, their delay times, the addition of delay time modulation and the amount of feedback used on the delay line, as well as other "processing" to the signal. The overall effect of changing these characteristics is to create a real or unreal reverberation or delay effect. Most units allow the user to modify the "parameters" of the program to obtain different delay times, modulation, feedback and processing. Often the units will also let the engineer store his/her settings as a user-programmed effect.
In the 1960's and during the "infant" period of multitrack recording, the engineer used one basic reverberation device or program to make the recording sound like it was recorded in an acoustic environment other than the studio. Studios tended to be made very "dead" and the instruments were close-miced to capture the direct sound with none of the studio sound. The reverberation device then added the "room" to the sound. To create a "very-real" recording, the band was taken to a concert hall or very large recording studio with desirable acoustics that simulated a concert hall.
With the availability of many reverberation devices, a new approach is often used in mixing. Today the engineer will tend to use a reverberation or delay program that enhances the particular instrument and freely use several reverberation programs in the mix. In addition, several reverberation programs will be blended for one instrument. It is not uncommon for a dozen different programs to be used in the same mix. It is uncommon for as few as three to be used.
The "Real" Effects:

Most digital delay devices will have programs that recreate reverberation sounds that have been used since the early 1960's.

Hall Programs:

Hall programs simulate the acoustic effects of a concert hall or auditorium. Hall programs are often used as the primary program in "smooth" acoustic-type instruments such as vocals, violins, flutes, etc. Hall programs, by their very nature, would be used for classical music. A characteristic of the hall program is a time delay before the reverberation hits, caused by the large dimensions of the concert hall.  In a reverberation program this is the pre-delay parameter.

Hall programs are also used as a secondary program for more percussive, instruments adding a natural ring-out ("tail") to the main reverb sound obtained with other reverberation programs. 
The pre- delay associated with the hall program means that the reverberation will not greatly effect the perceived placement of the instrument in the mix. Per Haas, the direct sound gives the listener a key to where the instrument is coming from. In a mix, adding reverberation causes an instrument to sound further away, but the pre-delay in the hall program lessens this effect dramatically. Using the hall program on a lead vocal and other programs on other instruments allows more reverberation to be used on the vocal while still allowing an "up-front" vocal sound.
If the reverberation hits too fast, the syllable of the lyrics can be covered up making the lyric hard to understand. "Hard" consonants such as begin to sound like other consents; for example a "ch" may be mis-recoginized as an "s" or distinguishing between "t" and "d" becomes more difficult.  The initial time delay of the hall program thus helps keep the lyrics understood. This same time delay causes a problem of a "double-hit" effect of percussive instruments such as drums; because of this, the hall program is a seldom-used program, by itself or as the primary reverb sound, for drums.
Figure1 shows the volume of the reverberation of a hall program over time. After the direct sound occurs, there are a few "early reflections" which are echoes caused by the sound bouncing off of one surface before reaching the listener. Somewhat later the reverberation is established. The time between the direct sound occurring and the reverberation hitting is called the "pre-delay" time on the reverberation unit.

Figure 1 - Hall Reverb Volume Envelope

Plate Programs:
Since the early 60's plate reverberation devices have been used. A plate reverb is a large (usually 4' by 8') plate on spring-clips that is vibrated in a similar manner to a speaker. One or two contact microphones (which convert the vibrations into an audio signal) are attached to the plate. Since many vibrations of the plate will occur, there is a reverberation created that simulates room reflections. Digital reverberation devices create a similar sound to the plate reverb on their "plate" programs.
Plate reverbs were a sensation when they came out - much more natural than artificial reverb devices that were available at the time.  By today's standards the sound is much less natural than other "real" reverb programs, but up to that point the best artificial reverb device was a spring reverb that was very "boingy" on percussion.  The spring reverb was one-dimensional, the concert hall natural reverb was three-dimensional and the new plate reverb was two-dimensional.  Plate reverb units sounded "great" when they were mechanically tuned - an arduous process process that took 4 hours of patience.  Modern digital reverb programs do a great job of simulating the sound of a well-tuned plate. 
Plate reverb programs tend to be less natural than other "real" reverberation. In the plate reverberation programs the reverberation starts immediately after the direct sound without a time delay.  Figure 2 shows the amplitude envelope of the reverberation signal obtained by a plate program.

Figure 2 - Plate Reverb Volume Envelope

The immediate reverberation of the plate program make it very useful for percussive instruments. The plate program does not have the "double-hit" effect of the hall program.

The lack of pre-delay also makes the plate reverb "cover up" the syllables of the lyrics, making them harder to understand. The lack of pre-delay will make a vocal sound more distant making it a useful program for background vocals and background instruments. Adding pre-delay to the plate reverb tends to make it closer to the hall program.

An acoustic echo chamber is built by building a room with non-parallel surfaces and then applying shellac to all surfaces to get them acoustically reflective.  A speaker and one or more microphones are installed in the room to pick up the reflections of the speaker's output.  This type of reverb was very popular in the 60's and was one of the basic elements of the famous "Motown Sound." Digital reverberation devices simulate this sound with Chamber or Echo Room programs.
Home recording engineers often used bathrooms as home chambers, but professional chambers are usually bigger and get much better results.
The characteristics of the reverberation are shown in Figure 3. The reverberation has, like the hall program, early reflections and a delay before the reverberation "hits." Because the room is much smaller, the early reflections and pre-delay are over much sooner than the Hall program. The shorter delay, usually around 10 ms, is long enough to keep vocals "up-front" sounding and short enough to prevent the "double-hit" sound on percussion. Because of these desirable characteristics, this program can be described as the "universal reverberation" - a good choice if only one program is available.

Figure 3 - Chamber Reverb Volume Envelope

The acoustic echo chamber or Chamber program also has a distinct disadvantage. The smaller room size of the chamber tends to give peaks and valleys in the midrange due to phase cancellation and standing waves. Acoustic chambers also tend to be "bright" (have extra high-frequency energy) because of the hard surfaces. These "colorations" tend to make this program less popular.
Other Natural Reverberation Programs:
Other "natural" reverberation and echo programs are available on modern reverberation devices.
Slap Echo (Delay Programs):
We discussed the slap echo when we were discussing long delay times. The program used in a reverberation device is usually called Delay L/R which gives two distinct delays (one left and one right) using different delay times.
Flutter Echo (Echo Programs):
Flutter echo occurs naturally between two parallel surfaces where the sound alternately bounces back and forth between the two surfaces. Empty rooms often have this characteristic. Flutter echo can be simulated with Echo L/R programs on reverberation devices. This is usually, however, used as a special effect rather than trying to simulate real acoustic environments.
Early Reflections:
Early reflection programs just generate the early reflections associated with halls and rooms. Using this program by itself gives an effect similar to slap echo. This program can be used before a Chamber or Plate program by patching the output of one device into the input of a second device; doing this makes these programs sound more like a Hall program.
The first programs made for digital reverberation units were the "natural" reverberation listed above.  A word of caution should be given to readers who listen to programs with the above names on inexpensive reverb units.  A good studio-quality reverberation "box" could cost $5000 but there will be boxes for a few hundred that have the same names as above - just not the same sound. 

Copyright 1993, 2000, 2001, Robert Dennis, ALL RIGHTS RESERVED