wp010adb09.png
The Sanskrit Alphabet
Since the Sanskrit devanagari alphabet has 49 letters, there are obviously problems when transliterating Sanskrit words into English. There are various methods, some of which simply go for the nearest sound in English, so that, for example, an English "s" can replace two or even three Sanskrit letters. Others, which are more accurate, put dots over or under certain letters, and little strokes (called "macrons") over other letters, and so on. It is now possible to reproduce this in some fonts, so I am gradually replacing the previous messy system I have been using.  This is the complete Sanskrit alphabet, in traditional order (vowels first, then consonants from guttural to labial, then semi-vowels, s’s and h):
a   ā   i   ī   u   ū   ŗ   ṝ   ḷ   ḹ   e   ai   o   au   ṁ (or ṃ)   ḥ
k   kh   g   gh   ṅ
c   ch    j    jh   ñ
ṭ    ṭh    ḍ   ḍh  ṇ
t    th    d   dh  n
p  ph    b   bh m
y    r     l      v
ṥ   ṣ     s
h
There is more information on the Omniglot website, if you are interested!

Elsewhere on the internet, many sites use the "Harvard-Kyoto" convention, which represents the Sanskrit alphabet by using capital letters to replace certain Sanskrit letters. For example, it is the one that is used for the “Online Sanskrit Dictionary”, and you need to know it to look words up.   The main problem with this system is that it makes familiar words look odd. For example Shankara appears as "ZaGkara"; ahamkara is "AhaGkara"; astanga is "ASTAGga"; Patanjali is " PataJjali"; shravana is "ZravaNa"; and so on.
The following table should help:

Harvard-Kyoto

Standard

Harvard-Kyoto

Standard

Harvard-Kyoto

Standard

A

long "a"( ā)

lRR

long "ul" (ḹ )

Th

retroflex "th" (ṭh)

I

long "i" (ī )

M

nasal "m" (ṃ)

D

retroflex "d" (ḍ)

U

long "u" (ū)

H

aspirate "ha" (ḥ)

Dh

retroflex "dh" (ḍh)

R

"ri" (ŗ)

G

guttural "n" (ṅ )

N

retroflex "n" (ṇ)

RR

long"ri" (ṝ)

J

palatal "n" (ñ)  

Z

palatal "sh" (ṥ)

lR

"ul" (ḷ)

T

retroflex "t" (ṭ)

S

retroflex "sh" (ṣ)