4.2.3 Prokaryotic Fossils: Isolated
or in Colonies
4.2.3.1 Criteria for the Identification
of Microfossils
The identification of microbial microfossils is tricky, but
taking great care and with well-preserved specimens, it is
possible (Westall and Cavalazzi 2011). The first difficulty
lies in the fact that non-biological structures can look strikingly like bacterial cells (Garcia-Ruiz et al. 2003) (Fig. 4.7)
and, thus, it is necessary to determine the biogenicity
(biological origin) of a particular biosignature. Moreover,
recent contamination by endolithic* microorganisms is also
possible. Thus, 8000 year-old cyanobacteria and filamentous
fungi, were identified in geological formations dating back
to 3.7 Ga (Westall and Folk 2003).
To reduce errors in the identification of microbial
microfossils (some of the controversies caused by misinterpretation will be discussed below), it is imperative not to rely
on a single biosignature, whether morphological or biochemical, because no individual signature is in itself sufficient evidence of biogenicity since many biosignatures can
be mimicked by abiogenic features. To validate the
biological origin of an observation, three types of studies
should be undertaken (Westall and Cavalazzi 2011):
1. Study of the geological setting at the macroscopic,
microscopic or elementary scales to determine whether
the environment in which the rocks formed could have
hosted life. It is also necessary to determine what changes
the rock may have undergone since its formation (diagenesis, metamorphism), which could alter the biosignatures.
2. Geochemical study of the proposed biosignature by measurement of carbon and sulphur isotopes, determination
of the presence of life-essential elements, such as carbon,
hydrogen, oxygen, nitrogen, sulphur and phosphorus, as
well as the presence of organic molecules showing compositional/structural complexity. For example, carbon
isotopes with δ
13 C values generally less than – 30 ‰
are typical of methanogenic archaea and those around –
27 ‰ of oxygenic photosynthetic bacteria.
3. Morphological study of the presumed microfossils. At the
cellular level, features such as size, shape, division or
lysis are sought. On a larger scale, colonies, biofilms
or mats should exhibit: (a) associations of structures
of the same size and shape that may represent cells of
the same species (colonies) or of structures of different
size and shape (formation of a consortium), (b) the presence of microbial exopolymers (extracellular polymeric
substances, referred to as EPS), and (c) structures with a
Box 4.1 (continued)
nitrogen with a
15 N/
14 N ratio ¼ 0.00368. The accuracy of an IRMS measurement is ~ 0.1–0.2‰ for
13 C/
12 C (ca. 5–7‰ for D/H). The differences between
the isotopic ratios of an element in a sample (R samp )
and in a standard (R std ) are very small; they are
expressed as ‘δ’ values, which record the enrichment
in units of ‰ relative to the standard according to the
equation:
δ ¼ R samp À R std
À
Á =R std
Â
à  1, 000
Thus, for carbon, this equation is expressed as
δ
13 C samp ¼
13 C=
12 C samp =
13 C=
12 C PDB
À 1
h
i
 1, 000
If the isotopic ratio of the sample is lower than that
of the standard (which is often the case, especially for
carbon), δ is negative. A sample with a δ value lower
than another sample is said to be ‘depleted’ in the
heavy isotope of the element considered. Conversely,
higher δ values are described as ‘enriched’.
The measurement of δ values in geological samples
represents a highly powerful tool (or ‘proxy’) that can
provide information on the origin (e.g. biological vs
abiotic) of the analysed element and can be used to
reconstruct its geological or biogeochemical ‘history’.
For example, a δ
13 C value of À50 to À130‰
in organic matter indicates formation through methanotrophic bacterial activity, while a δ
34 S value of
about À20‰ in diagenetic sulphides (e.g. pyrite)
signifies sulphate-reducing bacterial activity.
It is important to remember that the δ value of a
complex sample (or changes in δ values observed
through time) is not necessarily representative, since
multiple origins and processes can overlap and ‘hide’
(or modify) the original isotopic signature, thus limiting the interpretation of the measured bulk signal. This
is particularly the case for ancient biological samples,
whose stable isotopic composition could have been
altered over time by abiotic processes such as thermal
maturation, pressure and evaporation (Hoefs 2004).
In this kind of study, a ‘multi-proxy’ approach is
generally preferred.
4 For Three Billion Years, Microorganisms Were the Only Inhabitants of the Earth
87
or in Colonies
4.2.3.1 Criteria for the Identification
of Microfossils
The identification of microbial microfossils is tricky, but
taking great care and with well-preserved specimens, it is
possible (Westall and Cavalazzi 2011). The first difficulty
lies in the fact that non-biological structures can look strikingly like bacterial cells (Garcia-Ruiz et al. 2003) (Fig. 4.7)
and, thus, it is necessary to determine the biogenicity
(biological origin) of a particular biosignature. Moreover,
recent contamination by endolithic* microorganisms is also
possible. Thus, 8000 year-old cyanobacteria and filamentous
fungi, were identified in geological formations dating back
to 3.7 Ga (Westall and Folk 2003).
To reduce errors in the identification of microbial
microfossils (some of the controversies caused by misinterpretation will be discussed below), it is imperative not to rely
on a single biosignature, whether morphological or biochemical, because no individual signature is in itself sufficient evidence of biogenicity since many biosignatures can
be mimicked by abiogenic features. To validate the
biological origin of an observation, three types of studies
should be undertaken (Westall and Cavalazzi 2011):
1. Study of the geological setting at the macroscopic,
microscopic or elementary scales to determine whether
the environment in which the rocks formed could have
hosted life. It is also necessary to determine what changes
the rock may have undergone since its formation (diagenesis, metamorphism), which could alter the biosignatures.
2. Geochemical study of the proposed biosignature by measurement of carbon and sulphur isotopes, determination
of the presence of life-essential elements, such as carbon,
hydrogen, oxygen, nitrogen, sulphur and phosphorus, as
well as the presence of organic molecules showing compositional/structural complexity. For example, carbon
isotopes with δ
13 C values generally less than – 30 ‰
are typical of methanogenic archaea and those around –
27 ‰ of oxygenic photosynthetic bacteria.
3. Morphological study of the presumed microfossils. At the
cellular level, features such as size, shape, division or
lysis are sought. On a larger scale, colonies, biofilms
or mats should exhibit: (a) associations of structures
of the same size and shape that may represent cells of
the same species (colonies) or of structures of different
size and shape (formation of a consortium), (b) the presence of microbial exopolymers (extracellular polymeric
substances, referred to as EPS), and (c) structures with a
Box 4.1 (continued)
nitrogen with a
15 N/
14 N ratio ¼ 0.00368. The accuracy of an IRMS measurement is ~ 0.1–0.2‰ for
13 C/
12 C (ca. 5–7‰ for D/H). The differences between
the isotopic ratios of an element in a sample (R samp )
and in a standard (R std ) are very small; they are
expressed as ‘δ’ values, which record the enrichment
in units of ‰ relative to the standard according to the
equation:
δ ¼ R samp À R std
À
Á =R std
Â
à  1, 000
Thus, for carbon, this equation is expressed as
δ
13 C samp ¼
13 C=
12 C samp =
13 C=
12 C PDB
À 1
h
i
 1, 000
If the isotopic ratio of the sample is lower than that
of the standard (which is often the case, especially for
carbon), δ is negative. A sample with a δ value lower
than another sample is said to be ‘depleted’ in the
heavy isotope of the element considered. Conversely,
higher δ values are described as ‘enriched’.
The measurement of δ values in geological samples
represents a highly powerful tool (or ‘proxy’) that can
provide information on the origin (e.g. biological vs
abiotic) of the analysed element and can be used to
reconstruct its geological or biogeochemical ‘history’.
For example, a δ
13 C value of À50 to À130‰
in organic matter indicates formation through methanotrophic bacterial activity, while a δ
34 S value of
about À20‰ in diagenetic sulphides (e.g. pyrite)
signifies sulphate-reducing bacterial activity.
It is important to remember that the δ value of a
complex sample (or changes in δ values observed
through time) is not necessarily representative, since
multiple origins and processes can overlap and ‘hide’
(or modify) the original isotopic signature, thus limiting the interpretation of the measured bulk signal. This
is particularly the case for ancient biological samples,
whose stable isotopic composition could have been
altered over time by abiotic processes such as thermal
maturation, pressure and evaporation (Hoefs 2004).
In this kind of study, a ‘multi-proxy’ approach is
generally preferred.
4 For Three Billion Years, Microorganisms Were the Only Inhabitants of the Earth
87